SENATORIAL AMBIIIIIN AND LEGISLA'I’WEBEHAVIOR MICHIGAN STATEUNWERSITY. .fi . MURRAY FROST 1972 w . . LIBRARY W Midligan Scan: University This is to certify that the thesis entitled Lenatorini Anoition and Legislative Behavior presented by Furrly Frost has been accepted towards fulfillment of the requirements for Pl? . T). degree in P01 '1 't icril. Lolenco A- Q , . 44/037121 d . 04' 1&4 L‘ (R I v V K/ I Major professor DateJ'oh- 2/}; 1:17? “ 0-7539 l p—-—_...______. ...—.__.- %__ *mkm ~—..._ __.. 1—.“ 6..“ I III; IIIIIIII LIN III III I II IIIIII I ABSTRACT SENATORIAL AMBITION AfiD LEGISLATIVE BEHAVIOR By Murray Frost Several assumptions implicit in ambition theory were tested by examining the background and behavior of the 85 United States Representatives who sought Senatorial nomina- tions in the 1950-1968 period (referred to as RSTs). This research supports ambition theory on some accounts, but also challenges it on several accounts. Joseph Schlesinger's work on ambition theory assumed that opportunity is a major stimulus to ambition. A model was developed which views progressive ambition as a func- tion of (a) the opportunity structure and other aspects of the destination office (i.e. Senate), (b) characteristics of the situation at the origin (i.e. House), and (c) the characteristics of the individual or agent. The data for the 85 BSTs indicate that RST participation in Senatorial contests varied directly with the level of opportunity im- plicit in a classification of Senate situations based upon the party and re-election plans of the incumbent (e.g. the greatest likelihood for HST entry occurred in the Barty Succession situation of sharing the incumbent's party and the latter not being a candidate for re-election). This impact of opportunity upon progressive ambition was in contrast to the other two sets of variables in the model. RSTs could not be differentiated from other Representatives Murray Frost on the basis of the personal characteristics examined; and although several aspects of this House situation differed (e.g. RSTs came from larger and more competitive districts), these were not statistically significant--nor were there consistent differences in legislative behavior patterns. Schlesinger's development of opportunity structure rested upon the previous pattern of successful nominations. If his assumption were correct, one would expect that RSTs from state parties where Congress was a major penultimate office for Senatorial nominees would face less competition for the Senatorial nomination than other RSTs. But the data indicated that less than half of the RSTs came from such states (nor were RST opponents affected by this aspect of opportunity structure) and that the degree of competition for the nomination was not related to the pattern of nom- inees' penultimate office. A major assumption of ambition theory is that a legislator's behavior is a response to his office goals. This leads to the hypothesis that RSTs would change their behavior in their ultimate House term if it were inappro- priate for their prospective Senate constituencies. The data indicated that only 40% of the RSTs were Ready--i.e. already adhering to the appropriate strategy. But of the others. approximately two-thirds were Reluctants (i.e. did not change their behavior to conform) and only one-third were Responders. However, late-starters (RSTs whose candi- dacies occurred late in the term and only after a vacancy occurred)--omitted from this analysis--were even less Murray Frost likely to have changed their behavior in their last term, suggesting some impact of progressive ambition upon behavior. In summary, some of the findings supported ambition theory, but some did not. Since there was some evidence for the impact of opportunity upon progressive ambition, and for the impact of progressive ambition upon legislative behavior, the value of ambition theory as a heuristic model was not impaired. SENATORIAL AMBITION AND LEGISLATIVE BEHAVIOR BY Murray Frost A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Political Science 1972 2Q 31", COpyright by MURRAY FROST © 1972 ACKNOWLEDGKENTS The writer of any Ph.D. dissertation owes a debt of gratitude to his committee for their advice, encouragement, and helpful criticisms. But a dissertation effort stretched out over as many years as this one requires much more, and my acknowledgments must note these unusual debts. Of course I thank my committee: Prof. Joseph Schlesinger, Prof. LeHoy Ferguson, and Prof. Charles Press--not only for the usual contributions of a committee, but also for their patience combined with sufficient urging so that I was able to complete the project and maintain my sanity. I would also like to thank their doctors for maintaining their health long enough so that someone who started graduate school in 1957 could still have a committee who knew him from the beginning. Special thanks are due to my committee chairman, Prof. Schlesinger. His work served as a stimulus for this study, and his guidance was especially valuable. I am aware of the thanks I owe the Falk Foundation for their Fellowship awarded me in 1957-60 which enabled me to undertake and complete my graduate course-work. Although no one may recall that one of their requirements was a dissertation on American politics, I hereby acknowledge my debt to them as well as partially repay them with the 11 required research. In trying to list those contributing to the comple- tion of this work, I must thank my former employer General Research Corporation for their indirect incentive of including me along with the other Political Scientists in their employ as part of their contribution to President Nixon's war against inflation. My "reassignment" to the lines at the California Department of Human Resources Development provided me with the push I needed to finally complete "it". The last time I wrote an acknowledgments section, I publicly thanked my fiancee Marlen Diane Willis for her sacrifices. I would now like to thank publicly--and inade- quately--my wife Marlen Diane Willis Frost for her sacrifices. Her love and patience (and her "nudzhing") far outweighed everyone else's. Thengis one final group who must be acknowl- edged for their contribution-~I would like to publicly thank my children Michael, Mandy, and Marcy for playing outside, or in their rooms quietly, or watching TV in a hushed silence, and for willing to do without my attention for reasonably long periods of time over the span of their lifetimes. iii TABLE OF CONTENTS ACKNOWLEDGIVIEIJTSOO.OOOOOOOOOOOOOOO0....OOOOOOOOOOOOOOCC LIST OF TABLESOOOOCOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCO... Chapter I. IA'TRODUCTIOATOO0.......0OOOOOOOOOOOOOOOOOOOOOOO Senatorial Ambition Model................... N0m1nat10n Struggleoooooooo0.000000000000000 BehaVJ-OI‘ ChangeOOOOOOOOOOOOODOOOOOOOOOOOOO.0 Study's Population.......................... AGENT AND ORIGIN CHARACTERISTICS AND II. SEPJATORIAL AlfiBITIODIOOO0.0.00.0.0...00.00.... IntrOdUCtionooooooooooo.pogooooooggoooooogoo Agent CharaCtGTIStICS.ooaoooooooooooocoooooo Origin Characteristics...................... COnCIUSIOn.....o..o....................o.... III. DESTINATION SITUATION AND SENATORIAL AMBITION. Pllrpose.OOOOOOOCOOOOOOOOOOO00.0.00...0...... Senate Situation and Agent Characteristics.. Senate Situation and Origin Characteristics. Senate Situation and Other Destination CharaCteriSticsooooooooo00.000000000000000 COHCIUSIOnocoo0000000000000oooooooooooooooco IV. NOMINATION STRUGGLE ANALYSIS.................. Purpose and Hypotheses...................... FindingSOO0.0D...COO...COOOOOOOCOOOOOOOOOOOO Summary......o..................o........... Other Variables in the Nomination Struggle.. canolusioncoooooocoooooooooooooooooooooooooa V. STRATEGY-APTITUDE AND BEHAVIOR CHANGE......... IntrOdUCtiOnooooooooooooooooocoo-00.00.00... MethOdOlOgyooooooooooocoooococo-00.00.000.00 HYPOtheSeS..........o......o................ FindIHSSo0.0000000000000000...oooooooooooooo COHCIUSIOnSoooooo00000000000000.0000...coco. iv 11 vi 29 32 36 42 42 55 115 119 119 128 133 143 150 157 157 169 185 188 199 204 204 205 216 221 235 VIC CODICLUSIONoooooooo0000000000000000000000000000 25.3 APPENDIX: N MINATION STRUGGLE ACCOUNTS............... 265 BIBLIOGRAPHY.O.O.O.O.IOOOOOOOOOOOOOOOOOOOOOOOCOO...0.. 389 LIST OF TABLES Table 1. List of Representatives Who Tried for the Senate. 1950-19680000000000000000000.0000no. 2. Age at Entry Into First Public Office and House of Representatives, RSTs and Control Group (86th Congress)....................... 3. Median Age at Entry Into First Public Office, BSTS and CO-trOl Group..............oo.o.... b. Occupations of RSTs and House of Represen- tativeSOOO0..0....OOOOOOOOOOOOOOOOOOOOOOO... 5. Prior Office Experience of RSTs and Control GrOUPOOOOOOOOOOOO...OOOOOOOOOOOOOOOOOOOOOOOO 6. First Public Office of RSTs and Control Group. 7. Career Characteristics of "Recent R35" and "Earlier RSSHOOOOOOOOOOOOOOOOCCOOOOOOOOOOOO. 8. Education and Mobility of RSTs and Control Group...‘OOOOOOOOOOOOOOOO.OOOOOOOOOOOOOOOOOO 9. Proportionate Size of Congressional Districts (CDs) for RSTs and All Representatives...... 10. Size of City of Residence for RSTs and Control GrOUP.COO...DOODOOOOOOOOOOOOOCCOC0.0.0.0.... 11. House and Senate Seat Competitiveness for RSTs and ContrOl Groupoooooooooo00.000.000.900... 12. Terms of Service in the House for BSTs and All Representatives............................. 13. Freshman Committee Assignments, 80th to 90th Congresses.............o....o............... 14. RSTs' Ultimate House Committee Assignments and All House Committee Assignments............. 15. Index of Specialization and HST Success....... vi Page 39 45 46 #7 49 51 53 54 57 61 65 69 73 7Q 81 Table 16. 17. 18. 19. 20. 21. 22. 23. 2h. 25. 26. 27o 28. 29. Page Liberal-Conservative Rating of Terms of RSTs, State Party Medians, and All Representa- tives, 80th-90th Congresses................. 86 RST Liberalism-Conservatism in Relation to State Party Median for All Terms, 80th-90th congresses.OOOOOOOOOOOOOOOOOOOOOO000.000.... 87 Liberal-Conservative Average Rating for RSTs, 80th Congress to RST's Penultimate Congress. 89 Age, House District Competitiveness, and Senate Seat Competitiveness, and Liberal- Conservative Average Ratings for RSTs Compared to State Party Medians, 80th Congress to RST's Penultimate Congress...... 92 Age and House District Competitiveness, and Liberal Conservative Average Ratings for RSTs, 80th Congress to RST's Penultimate congressO0.00...0.00000000000000000000000000 95 Party Cohesion Level of Terms of RSTs, and Their State Party Medians, 80th-90th congresseSOOOOOO0.0.000000000000000000000000 101 RST Party Cohesion in Relation to State Party Median for All Terms, 80th-90th Congresses.. 103 Party Cohesion Average Rating for RSTs, 80th Congress to RST's Penultimate Congress...... 10b Party Cohesion Ratings for RSTs Compared to State Party Medians (80th Congress to RST's Penultimate Congress), and House District Competitiveness and Senate Seat Competitive- ness.‘0.0.00..0.0.0....00.000.000.000...0.0. 106 Party-Orientation Based on Relative Party Cohesion and Liberal-Conservative Ratings... 110 Seniority, Committee Assignment. and Party orientationOOOOOOOOOODOOOIOOOOOOOOOOOOOOOOOO 113 RSTs Classified by Senate-Incumbent's candidacy and Partyocooooooo0000000000000... 12“ Senate Situation and Age at Political Career P01nts.000......COO...OOOOOOOOOOOOOOOOOOOOOO 130 Senate Situation and Mean and Median Age at Political Career Points..................... 130 vii Table 30. 31. 32. 33. 340 35. 36. 37. 38. 39- #0. A1. #2. U3. #4, 45. 46. “7. A8. 49. 50. Senate Situation and First Public Office...... Senate Situation and Occupation............... Senate Situation and Prior Public Office Experience...OOOOOOOOOOOOOOOOOOO0.0.0.000... Senate Situation and Size of City of Resi- denoe, by Party.........o................... Senate Situation and Proportionate Size of Congressional District...................... Senate Situation and House Seniority.......... Senate Situation and Committee Assignment, by SeniorityOOOOOOO0.0..OOOOOOOOOOOOOOOOOOOOOO. Senate Situation and Index of Specialization (PenUItimate Term)oncocoa-0.0000000000000000 Senate Situation and Liberalism-Conservatism.. Senate Situation and Party Cohesion........... Senate Situation and Penultimate Office Pattern for Senate Nominees (State Party Opportunity Structure)...................... Senate Situation and Senate and House Competitiveness...OOOOOOOOOOOOOOOOOOOOOOO000 Senate Situation and Nomination Contest Competition...0.0.0.0000...OOOOOOOOOOOOOOOO. Previous Public Office Experience and Nomina- tion Struggle (NS)0.00000000000000000.000000 Occupation and NS............................. Place of College Education and NS............. Party Cohesion and KS......................... Specialization (Penultimate Term) and NS...... Size of City of Residence and NS.............. Congressional District Size and NS............ State Party Career Pattern (Penultimate Office) and NSC.O..OOOOOOOOOOOOOOOOOOOOOOOCO viii Page 131 132 133 134 135 136 137 139 141 1M3 144 146 1A8 170 171 171 172 17k 176 176 178 Table 51. 520 53. 5a. 55. 56. 57. 58. 59. 60. 61. 62. 63. State Party Career Pattern for Opposition and NSOOOCOOO0.00.0000...0.0.0.0000....0...O Source of Competition (Penultimate Office) and NS.......................................... Senate Situation and HS....................... Senate Competition and NS..................... House and Senate Competitiveness and NS....... District Size and Strategy Aptitude-Behavior Change (SABC)OOOOOOOOOOOOOOOO000.00.00.00.00 Senate Competitiveness and SABC............... Relative Size of City of Residence and SABC for Democrats............................... State Party Career Pattern and SABC........... Occupation and SABC........................... Age At Senate Attempt and SABC................ Political Experience and SABC................. Highest Committee Assignment and SABC......... ix Page 180 181 182 183 185 223 227 228 230 231 233 234 235 CHAPTER I INTRODUCTION In November 1970, New York City Mayor John Lindsay denied reports he might be a Presidential candidate saying, "It's not viable politically...because I am an enrolled Republican and also because it's unheard of for a mayor at any time to run nationally. It's never been done in the United States and people don't change very easily."1 Despite such a strong statement of the relationship between opportu- nity structure. ambition, and careers. this area has been studied by comparatively few political scientists. The relationship between these variables and political behavior has not been studied extensively either. This study focuses upon several of these relation- ships for a select group of American politicians. Those examined in this study are the 85 members of the House of Representatives who made attempts to become members of the Senate while still serving in the lower house, during the two decades (ten election years) from 1950 to 1968. Any attempt to learn about the achievement of this "progressive ambition"2 must include the unsuccessful-~i.e. those who 1Los Angeles Times. November 10. 1970, Pt. I. p. 11. 2See Joseph A. Schlesinger, Ambition and Politics: gglitical Careers in the United States (Chicago: Rand McNally & Company, 1966). p. ID Tor the definition of this term. 1 2 either lost the election or who lost the struggle for the nomination--and not Just those who were successful. The major stimulus for this study has been the work of Joseph Schlesinger focusing upon ambition and the attain- ment of public office. In his major study, Ambition and Politics. he defines the structures of opportunity in the various states (and national government) for the major offices. In other words he begins by tracing the careers of Senators and Governors and defeated candidates for these two offices. His major theoretical contribution is to relate this to party theory. discussing both the nature of party organization as well as the degree of convergence of the parties in their behavior. ' Although his theoretical and empirical work revolves around the progressive ambitions of officeholders, his con- cern with who will want to advance involves the assumption that the opportunity structure in a state will excite or restrain a man's ambition. "The most reasonable assumption is that ambition for office, like most other ambitions, deve- lops with a specific situation. that it is a response to the possibilities which lie before the politician."3 He recog- nizes that there are factors other than the opportunity structure which affects ambition--for instance a man's age may serve to modify his ambitions. Our conception of the forces giving rise to progressive ambitions for the Senate is broader. The "specific situation" involves both his 3Ibid.. p. 8. 3 current office and the one to which he aspires, as well as personal factors. Schlesinger notes that "In ambition theory the emphasis in analyzing such data is not on Egg one advances in politics but on who will want to advance."1+ But his data on who wants to advance is drawn from the careers of those who succeeded in their ambition. or who were nominees. The opportunity structure is shaped by the successes and failures of those seeking the nomination. This study, therefore, examines the accounts of the nomination struggles of these Senatorial candidates that constitute the population. The relationship of ambition to behavior is perhaps the underlying concern of Schlesinger. "The central assump- tion of ambition theory is that a politician's behavior is a response to his office goals."5 But Schlesinger's discussion is a theoretical one without any reference to data. He sug- gests the need to examine the interaction of strategies of advancement and opportunities in order to explain legislative or political behavior. But Ambition and Politics--as he admits~-neglects this in order to fully develop the structure of opportunity. The study reported here will examine the impact of Senatorial candidacy upon their behavior in the lower house. This study, therefore, centers about three broad questions. 1) In Chapters II and III we ask: How were “Ibid.. p. 15; emphasis in the original. 51bid., p. 6. 4 these Representatives who tried to become Senators6 different from, or similar to. other members of the lower house? In other words, do Representatives with progressive ambition for the Senate differ from their colleagues with either discrete or static ambition (i.e. content to continue in the House for the remainder of the term or longer)? Did they have dif- ferent backgrounds, careers, constituencies, patterns of legislative bahavior? 2) In Chapter IV we focus on: What was the nature of their Senate nomination contests? For example, how many faced serious competition for the nomination? Did RSTs in state parties where Congress was the predominant penultimate office for the Senate nomination face less oppo- sition than other RSTs to their attempted "promotion"?7 3) In Chapter V we examine: What effect does candidacy for the Senate position have upon the behavior of the legislator while he still is in the House? For instance, does his roll call behavior or his degree of specialization change? Senatorial_§mbition Model The "model" underlying the analysis in Chapters II and III says that the decision by a member of the House of Representatives to become a Senator is a function of three sets of variables. One set revolves around the "agent"-- the individual's personal characteristics and background. 6These are referred to as RSTs--Representative-to- §enator€§ryers~~throughout this study. 7See Herbert Jacobs. "Initial Recruitment of Elected Officials in the U.S.--A Model". Journal of Politics, XXIV (November, 1962), pp. 703-16, for the useful differentiation of "initial recruitment" and "political promotion". 5 The second set concerns the contest or environment at the "origin"--or House of Representatives. The third set con- cerns the situation at the "destination"--or Senate. Although the model may use the language of transportation analysis-- e.g. origin and destination--it can be reconciled with other ”models“ concerning recruitment. For instance. Barber8 noted that the literature con- cerning the relationship between political recruitment and official behavior has been studied in three ways. Some have focused upon political variables such as the political organi- zation and degree of competition. Other studies have concen- trated on sociological variables such as the socio-economic background of the decision-makers and candidates. While still other studies have concentrated on psychological variables, primarily the motivational characteristics of politicians. He then combined aspects of all three, and suggests that recruitment patterns are a function of moti- vation ("Do I want it"). resources ("Can I do it"), and opportunity ("Do they want me"). Our model. too, suggests that there are "agent" or personal elements-~such as occupation, previous public office experience, and other aspects of political careers (such as age). Motivation and other personality variables could be included here. This study. however, involved only published data and involved no interviewing which would be helpful if 8James David Barber, The Lawmakers: Recruitment and Adaptation to Legislative Life (New Haven: Yale University Press, 1965). 6 not essential in ascertaining this subset of variables, and they therefore are omitted. Our model also suggests that the level of resources available is important in determining whether a Representative will try for the Senate. Many of these resources may stem from his situation in the House. The nature of his House constituency-~e.g. the size of his district--may aid his Senate candidacy. His seniority, committee assignments, and specialization may be resources in themselves or be important assets with which to attract resources. Similarly his legis- lative behavior-—i.e. his voting record-—may be a resource to gain still other needed resources such as money, workers, and votes. Our model makes reference to opportunity too; it is viewed as part of the environment at the destination-«the Senate situation. Certainly which party holds the Senate seat and what the re-election plans of the incumbent are, are two important factors which will affect a Representative's decision to seek the Senate seat. The degree of competition for the seat is another important factor--a competitive situ- ation may be assumed to be less attractive to a legislator than the prospects of an easy victory in a safe state. Similarly the prospects of competition for the nomination may influence his decision. The previous pattern of public office holding of Senatorial candidates is another aspect of Opportunity. It may make him the object of Speculation con- cerning the office, and/or it may affect his ambition-~which is one aspect of his motivation-~and/or the ambition of 7 other officeholders or prOSpective candidates. It is these three sets of variables--l) agent or personal characteristics. 2) origin or House-related char- acteristics, and 3) destination or Senate situation factors-- that determine whether a member of the House of Representatives will seek promotion to the Senate. A. Personal Characteristics Very few studies have examined the characteristics of office-holders with progressive ambition. Although Schlesinger develops the concept of progressive ambition and differentiates this desire for higher office from the desire to be re-elected (static ambition) and the intention of holding an office until the end of the term (discrete ambition), he does not present any data. His theoretical discussion assumes that the opportunity structure gives rise to ambition. Soule is one of the few authors who have discussed the characteristics of a set of progressively-ambitious officeholders. His published study of members of the Michi- gan state House of Representatives, however, is largely concerned with political socialization--e.g. time and source of political socialization, formal education, and certain psychological traits or values.9 9John w. Soule. "Future Political Ambitions and the Behavior of Incumbent State Legislators", Midwest_Journal_of ggiitioa1_seience. XIII (August, 1969). pp. £39-54. Also see his Ph.D. dissertation cited there: John w. Sonic, ”The Influence of Political Socialization, Interpersonal Values and Differential Recruitment Patterns on Legislative Adaption: The Michigan House of Representatives" (unpublished 8 One of the variables that Schlesinger examined for its impact upon ambition was age. "The younger a man is when he enters politics the greater the range of his ambition and the likelihood of his developing a career commitment.”10 Age at entry into public office, therefore, could be used as an indicator of progressive ambition. We would hypothesize. therefore. that Representatives with progressive ambition for the Senate. will have entered public office at a younger age than their colleagues in the House. An alternate hypothesis, however. is also suggested by Schlesinger.11 It is possible that late starters may use the advantages gained before entering public office to aid their rapid promotion. His data suggest that earlier starters do not necessarily gain major nominations (i.e. Senator or Governor) earlier. The earlier starters may be more career oriented and appear to have more years of public office experience before seeking major office. Another variable that many authors have examined is that of occupation. Jacobs called it the crucial social variable in his model of initial recruitment.12 Most studies of the social backgrounds of decision-makers‘have noted the over-representation of lawyers in legislative bodies. They have offered a variety of reasons for this. Jacobs places Ph.D. dissertation. Department of Political Science, Uni- versity of Kentucky. 1967). O H 92. Oltop p. 176. llIbid.. p. 182. 1292, Egg.. p. 709. 9 the most emphasis upon the lawyer's role of broker: he fre- quently is engaged in bargaining with outsiders and exchanging ideas. But he also notes other aspects of his occupation-- e.g. that it leaves him time to seek office. that he is already at the fringe of politics often contacting govern- mental officials. The lawyer's skills of verbal manipulation, comprehension of complex and technical information, and his experience in human relations have been noted by some, in addition to his bargaining skills. Others have speculated about the prestige and class status of the legal profession. Still others have commented upon the blending of the legal occupation and a political career in terms of the flexibility of his work schedule. the relatively slow change in the pro- fession enabling a defeated lawyer to return easily to his practice, and the aid that a political career can give to his legal career and income (even a political defeat may serve as advertising for his occupation). Schlesinger suggests the over-representation of lawyers among Governors and legislators may be due to their monopoly of public offices relating to the administration of law through the court system.13 All of these explanations for the over—representa- tion of the lawyer in legislative bodies are plausible, and there are some data to support each of them. But once a lawyer has reached the legislature, is he more likely to have a progressive ambition (e.g. for the Senate) than other legislators? Some of the reasons noted earlier may relate to 13See, Joseph Schlesingen,"Lawyers and American Politics: A Clarified View," Midwest Journal of Political W. I (May 1957). pp. 26-39. 10 political promotion as well as initial recruitment. For instance we noted the speculation about the ability of a lawyer to return to his legal practice if defeated, as a spur to a legislative career; but it also can serve to minimize the risks of seeking higher office. In other words, there are reasons to expect lawyers in the House of Representatives to be more likely than non-lawyers to have progressive ambi- tion for the Senate. There is some indirect evidence of the lawyer- legislator's ambition. For instance, Jewell and Patterson note that "Senates in the United States tend to contain a higher proportion of lawyer-legislators than do houses of representatives."14 And there are more lawyers in the national House of Representatives than there are in state legislatures. Zeller's study of state legislatures in 1949 indicated 19% of those serving in the lower house were lawyers compared to 33% of those in the upper house.15 A recent survey-~of the 89th Congress--found lawyers to be 55% of the House and 67% of the Senate. More directly, Eulau and Sprague note that in the four state legislatures surveyed by the State Legislative Research Project, lawyers were more likely than non-lawyers to acknowledge "their ambitions as reasons for seeking l""Malcolm E. Jewell and Samuel 0. Patterson, The Le islative Process in the United States (New York: Random House. 19535. p. 107. H lSBelle Zeller (ed.). American State Legislatures, (New Yorki‘ Thomas I. Crowell Cbmpany, 1954). p. 71. ll legislative office."16 They begin their political careers at an earlier age, and this may be assumed to indicate a greater degree of ambition. And it is relevant to note that the lawyer-legislators in that study were more likely than non-lawyers to be interested in politics at an earlier age-- 9.3. in each of the states over 70% of the lawyer-legislators for whom data on this question was ascertained were interested in politics at the time of college or earlier (the range was 72% to 84%) compared to 44% to 59% for non-lawyers.17 Early political socialization was found to be related to progressive ambition by Soule.18 This data suggests that the lawyer—leg- islator's ambition preceded his choice of occupation. Never- theless, occupation of the Representative can be used as an indicator of progressive ambition. We would hypothesize, therefore, that more RSTs than other Representatives will be lawyers. Studies of careers have shown that there are many paths to an office. For instance, when Schlesinger traced_ career paths using such minimal landmarks as first office, office experience. and penultimate office (by types--e.g. statewide elective. or law enforcement), he found over 30 such paths leading to the Senate with no single route followed 16Heinz Eulau and John D. Sprague, Lawyers in Politics: A Study in Professignal Convergence (Indianapolis: The Bobbs- Merrill Co.. 19641: Do 77- 171bid.. recalculation of Table 3.1. p. 57. 189p. 913.. p. 446 . 12 by as much as 10%.19 But as his study noted there were linkages between offices-~e.g. manifest functional relation- ships such as legislative offices, and the sharing of similar electorates. Of course all Representatives and Senators share the same function of legislator. and therefore all of the men in this study have this office experience in common. But one may suppose that earlier career paths may be related to a Representative's seeking Senatorial office. Therefore one may hypothesize that the first office, and public office experience of RSTs will differ from that of other Representatives. For instance. Schlesinger concluded, “The higher the office. the more complex the typical career leading to that office."20 Therefore since the Senate is a higher office than the House, it is reasonable to hypothesize that Representatives who have Senate ambitions will have more com- plex careers than Representatives without such ambitions. Similarly we may examine whether those who are successful in their attempt to move to the Senate have different career patterns than those who lose or those who are unsuccessful in their attempts to gain the nomination. Schlesinger's data for all Senators and all defeated Senatorial candidates (not just those whose penultimate office was the House) indicated 21 more complex careers for winners than losers. Mezey suggests that this may be a function of majority-minority party 19gmbition and_eg;itics. p. 92. 20Ibid.. p. 196. 211bid.. p. 96. 13 status.22 Complexity was measured by the difference between use of an office as a first office and as a penultimate office. Individuals whose careers include more offices may be said to have a more complex career than those with no or only one prior public office. Complexity may also be measured by different kinds of office experience. A Representative who only has had legislative experience may be said to have a less complex career than one who has had either law enforcement or admin- istrative experience prior to his election to the House. Therefore it is reasonable to hypothesize that RSTs more than other Representatives would have had law enforcement or administrative experience. Since a politician's career time is limited, it is reasonable to assume that if law enforcement or administra- tive experience is expected, that these men will start their public office career in these offices. Therefore we could hypothesize that RSTs more than other Representatives will have used these offices as first offices.23 22Michael L. Mezey. "Ambition Theory and the Office of Congressman." Journal of Politics, XXXII (August, 1970). ppc 563-790 23But other contradictory hypotheses are also possible --for instance if we assume a planned career, our study focuses upon a group of men who felt they ought to serve in the lower House before trying to get to the Senate: it may be reasonable to assume that these men might also feel they ought to serve in the state legislature before attaining national legislative office. Therefore state legislative office as a first public office could be hypothesized to be more common among RSTs than other Representatives. 14 Another logical framework from which to hypothesize, may center about the idea of deviancy. Since only a small proportion of Representatives try to become Senators, they may be viewed as a deviant group. If their Senate ambition is a sign of deviancy. perhaps their earlier career will also deviate from the norm. Therefore we would expect the careers of RSTs--whether it be first public office. or office experience-oto differ from other Representatives. One other agent characteristic--if. and where. the HST attended college-~will be investigated (although still other personal variables--such as personality traits-~would be theoretically relevant). Chroniclers of decision-makers' backgrounds have noted that the latter have a higher level of education than the general populace. Soule's study indi- cated that Michigan state legislators with a college degree were more likely to have progressive ambition than legisla- tors with less education.2u Apparently legislators with more education are more likely to be confident of their ability to handle higher office. Since college was an important political socialization agent for state legislators, the location of that education can be important in their career (and behavior). For example, Vines demonstrated the relationship between place of law school education and the civil rights record of judges in the Fifth Circuit.25 Although a college education may be broadening 2"'02,cit.. p. 447. 25Kenneth Vines, "Federal District Judges and Race Relations Cases in the South," Journal of Politics. XXVI (May. 1964). p. 351. 15 for all, attending a college in another state should result in a less local or parochial outlook. This reduced localism may breed a broader outlook and ambition. Therefore we would hypothesize that l) RSTs would be more likely than other Rep- resentatives to have attended a college, and 2) RSTs would be more likely than other Representatives to have attended an out-of-state college. B. House Situation The second set of factors affecting the decision of a Representative to become a Senatorial candidate centers around the nature of his situation in the House. This includes dis- trict characteristics such as the size and competitiveness of the district. aSpects of his career in the House such as his seniority and committee assignments, and aspects of his beha- vior including specialization, party cohesion, and liberal- conservatism. The importance of these factors stems. in part, from their effect upon the level of resources which a Repre- sentative can acquire for use in his attempt to gain the Senate seat. For instance, Schlesinger observed, "When two offices have the same or similar electorates, it is logical to expect movement from one to the other."26 We would hypothesize, therefore. that a disproportionately large number of RSTs were elected to the House of Representatives in At-Large elections. These men have already faced a statewide electorate success- fully. They have already put together a party nucleus--a N 26Ambition and Politics. p. 99. 16 team of advisors and workers-~who have been able to solicit enough money, endorsements, and other forms of support to win a plurality of the votes from a statewide electorate. In general. we may hypothesize that the larger the share of the Senate constituency that falls in a Representative's district, the more likely he is to run for the upper house. He is also more likely to run for the Senate seat if the degree of competitiveness of his two constituencies--i.e. his current one in the House and his prospective one for the Senate-~are similar. Schlesinger has suggested that nuclear party organizations and officeholders in competitive situations are likely to adopt similar strategies of trying to minimize party differences by adopting moderate policies in order to appeal to that portion of the electorate which can become the margin of victory by swinging to either party.27 In contrast to this moderating tendency of competitive parties, he sug- gests that dominant parties will try to emphasize party differences. In either case, a Representative who faces the same needs in both his House and Senate situations will be in a better position to press for the Senate seat, than a Repre- sentative whose House and Senate needs differ. The "organi- zational tensions" of the latter situation may inhibit some Representatives' progressive ambitions for the Senate, in contrast to the former situation where the most appropriate Pattern of behavior in the House also is the most appropriate for his Senate situation. We would hypothesize, therefore, m 27Ibid., see Chapter 7. 17 that RSTs are more likely than other Representatives to come from situations in which the competition of their House dis- trict is similar to that of their states' Senate seat. We can also expect to find that more of the unsuccessful tryers for the Senate nomination (Ts) than successful nominees will be from "organizational tension" situations. Similarly. those RSTs who win the Senate seat can be hypothesized to be more likely than losers to have similar House and Senate competitive situations.28 The city in which the Representative lives may be con~ sidered a resource or at least an aid to acquire resources. If we assume that a Representative may be better known in his home town than elsewhere in his district or the state, then Representatives from larger cities should be at an advantage when seeking the Senatorial nomination. But Matthews has pointed out that large cities, as the birthplace of Senators, were under-represented, while small towns and cities were over- represented. He suggests that a small town background may be advantageous because it combines aspects of both rural and urban life. He also notes that perhaps the small town resi- dent "is exposed to a cross-section of the community. to those 'above' and 'below' his position on the social ladder" in con- trast to the isolated rural resident or the segmented urbanite.29 28A discussion of the relationship of the absolute-- rather than the relative--degree of competitiveness to pro- gressive ambition for the Senate appears later on pp. U 29Donald R. Matthews, S Senators and Their World (New York: Vintage Books, 1960). p. 16. ‘L 3“ 18 v.0. Key's analysis concerning the effect of primaries upon state political systems suggests a more refined hypothe- sis. Key found that because of the differential turnout of one-party areas in the dominant party's primary, that a dis- proportionate number of candidates for state office come from the party's safe areas. For instance. he reported, that in the decade or so prior to the primary system in Massachusetts 25% of the Democratic nominees for statewide office came from Boston and vicinity. and 40% of the Republicans came from Boston. In the first four decades of primaries there. 59% of the Democrats came from the Boston area, while the proportion of Republicans declined to 25%.30 This analysis suggests that Democrats from large cities--which are usually the stronghold of Democrats--would be at an advantage compared to Democrats from other areas. Similarly. rural areas and small towns and cities are more likely to be G.0.P. strongholds, and Republi- cans from these areas would be at an advantage compared to Republicans from larger cities. Our hypothesis is that these assumed advantages would lead to an over-representation of these types among RSTs when compared to other Representatives. As the Representative's career progresses in the House --i.e. as he gains seniority-~his position in that chamber improves. He becomes more knowledgeable concerning the issues, Procedures, as well as other legislators and actors in the legislative process. This places him in a stronger position to gain the resources necessary to run a successful Senate m 30v.0. Key. Jr., American Stat; Politics:tAn Intro- éssiiea (New York: Alfred A. Knopf. 1956?. pp. 155-6. 3" 19 campaign. His higher seniority may enable him to gain more important committee assignments, which may also help him fulfill his Senate ambition. This speculation would lead us to hypothesize that a higher proportion of RSTs than Representatives without pro- gressive ambition serve on the committees considered to be more important. Similarly, if seniority improves a Representative's ability to amass the resources necessary for a Senate contest, then we can assume that these Representatives will be more likely to have progressive ambition and we would expect a higher proportion of RSTs than other Representatives to have high seniority. But on the other hand. if a Representative has sufficient power in the House, he may have no interest in any other office. Therefore a more realistic hypothesis would be to suggest a curvilinear relationship between seniority and progressive ambition. Specifically we would hypothesize that RSTs would be less likely than other Representatives to be either newcomers to the House or among its more senior members. The relationship between a Representative's progres- sive ambition for the Senate and his legislative behavior will be developed further later, and will be the subject of empirical testing in Chapter V of this work. But at this point the linkage involves such questions as: Do RSTs vote more often with their party than do other Representatives? Are they more likely to be liberals or conservatives? Are they more likely to be generalists or specialists? Our model so far suggests the situation in the House --including legislative behavior-~18 a factor for the w IV ‘ Q‘IQ 20 Representative to consider in determining whether to try for the Senate. Our initial concern focuses upon the relation- ship of legislative behavior as a resource or an aid to gaining the resources necessary to make an attempt to gain the Senate seat. Obviously the situation at the Senate level is crucial: there can be no single national answer about the utility of a liberal record in gaining a Senate seat (even though it may be possible to argue that a candidate with a statewide consti- tuency must be more receptive to the more "liberal" demands of the urban interests for governmental services). Nevertheless, the initial Speculation offered here will be without reference to the Senate situation. Specialization is one of the norms of behavior in many legislatures.31 Specialization may evolve from the legislator's background-~an educator turned legislator may focus his atten- tion upon the school problem. or a retired army officer may consider military affairs his specialty as did one of the RSTs in this study (Rep. Frank Kowalski). Or it may come from prior specialization in a state legislature. Or it may spring from one's committee assignments--a member of the Ways and Means Committee may specialize in tax legislation. Although com- mittee assignments in Congress often benefit a legislator's relationship with his district--i.e. an argument that a particular committee assignment is important for the district and for re-election will be given prime consideration by the M 31For example. see Matthews, op.cit,. pp. 95-7: John C. Wahlke. et.al.. The Legislative System: Explorations in Le islative Behavior (New York: John Wiley and Sons, Inc.. 1932). chapter 9. ' 21 Committees on Committees32--the relationship between district and committee assignments is far from perfect. It is not un- likely. therefore. that a legislator's degree of specialization or area of specialization is not related to his ambition-- either his immediate static ambition or his future progressive ambition. It is possible to assume, then, that the generalist's role may fit the needs of a legislator with progressive ambi- tion better than would the specialist's role. A generalist may be better able to meet the interests and demands of a new constituency better than a specialist (especially if the characteristics of the districts differed). In addition, a generalist would be better able to adjust to a new committee assignment than would a specialist. Our speculation. there- fore, is that if a legislator is a generalist. his perspective will be broader and he would be more likely to think about future progressive ambitions for the Senate. The "model" would lead us to hypothesize that RSTs would be less likely to be specialists than other Representatives. But in view of the labor involved in calculating an Index of Specialization for a matched sample of Representatives who could serve as a con- trol group, and because there are no previously published findings for the House of Representatives using the measure of specialization developed by Matthews, this hypothesis will not be tested in this study. Instead, we speculate that an RST's level of specialization is related to his degree of 32See Nicholas A. Masters. "Committee Assignments in the House of Representatives", aperican_3olitical Science Review, LV (June, 1961). pp. 345-57. 22 success in satisfying his progressive ambition for the Senate. We would hypothesize, therefore. that Specialization is inver- sely related to success; we would expect to find more genera- lists among those who won their attempts to move to the Senate than those who lost, and we would expect to find the least proportion of generalists among those who were unsuccessful in even gaining the Senate nomination (the Ts). It is difficult to consider a legislator's degree of liberalism as a resource without further knowledge about the composition or make-up of his statewide constituency. But Soule's study of progressive ambition among the Michigan state legislators found "a small association between liberal- conservatism as an ideology and political ambition."33 It may be appropriate to speculate that those who see a more positive role for government in the solution of problems (i.e. a more liberal ideology) may have progressive ambition for the Senate. The Senate is considered to be the more liberal of the two branches,3u andzaliberal in the House may feel that he would feel more comfortable and be able to accomplish more if he were in the Senate. In addition a liberal stance may be assumed to be more receptive to urban interests, and so a Representative may view his liberal record as an important resource in any attempt to be elected from a statewide con- stituency. Having this resource-—i.e. a liberal voting 3322;2i3;. p. 453. He found gamma = .12, .10, 3“Lewis A. Froman, Jr.. Con ressmen and Their Con- stituencies (Chicago: Rand McNally & Co.. 1963), Chapter 23 record--may contribute to his decision to become a Senate candidate. Therefore the hypothesis to be tested at this point in the study is: RSTs are more likely than other Representatives to be liberals. In Chapter V of this study, this hypothesis will be refined to take the Senate situation into consideration. If we were to treat party cohesion independently from liberal-conservatism--despite their high interrelation which has been commented upon by many observers35--our model to this point does not suggest any clear relationship between party cohesion and the decision to seek Senate office. It is pos- sible to argue that those with progressive ambition will be more likely than the average Representative of his state and party to show high party cohesion because they would want party support for their future race, which might be more likely to go to a loyal party man than a maverick.36 On the other hand, it is logical to assume that the statewide constituency is approximated best by the average constituency, and therefore the party cohesion level of the average legislator in the state party would be the level of cohesion of those with progressive ambition. But it may also be possible to Speculate that a 35l/iatthews,op.cit.. p. 133: W. Wayne Shannon. Pagpy, Congtituencngnd Conggessional Voting:__A_§tudy of_Legislative Behavior in the United States House of Representatives (Baton Rouge: Louisiana State University Press. 19687, Ch.IV. 36See. Murray Frost, "Roll Call Cohesion in the New Jersey State Legislature. 1956-1957". (Unpublished M.A. Thesis, Michigan‘State University. 1959). for an indication of higher cohesion among those with future political ambition. But it should be noted that New Jersey had a very strong party system at that time, and even then the finding was not statis- tically significant. 24 candidate for the Senate would have to appeal to more or dif- ferent voters than a Representative and that a record of independence (i.e. low party cohesion) might be more fruit- ful for a Senate candidate. There is some data in Soule, for example, that notes that those with progressive ambition were more likely than other state legislators to be Trustees. Trustees are more independent and therefore are less likely to be voting with their party.37 In this portion of the study. then, we are interested in examining the party cohesion level of Representatives with progressive ambitions for the Senate to see if there are any patterns that are independent of state situations, even though the model at this point would not lead us to expect one rela- tionship more than another. We have just related several aspects of the situation at the so-called "origin"--i.e. the House of Representatives-- to the availability of. or ability to gain, resources, which in turn would affect a Representative's decision to seek the Senate seat. We hypothesized. for instance. that Representa- tives whose House district's competitiveness matched that of their Senate seat's would be more likely to run for higher office. Similarly. we felt that power and resources increased with better committee assignments and seniority, but that too much power in the House might inhibit progressive ambitions. This suggests that the decision to become a candidate 37See, Roger H. Davidson. The Role of the Congressman (New York: Pegasus, 1969). p. 161, for his finding that Delegates expressed more party loyalty than Trustees. 25 for a higher office (i.e. progressive ambition) is not only a function of the resources available (and of the candidate's personal characteristics-~already discussed-~and the opportu- nity for office-~to be discussed) but also of the political risk involved. Seligman's model sees the "role of parties in the recruitment of political leadership...as the product of the inter-action between the internal organization dynamics and diverse party environments as mediated by the structure of political opportunity and political risk."38 Effective poli- tical opportunity refers to conditions, motivations, and resources present or available to the prospective candidate. These are similar to the three sets of variables in our model. "Political risk refers to the extent of the losses in status, income. and influence when a person loses political office."39 One of Seligman's associates defined political risk in terms of the "candidates' perception of the costs and benefits associated with candidacy..."l+O Risk is not only a function of the office he seeks, but of the one he holds now. If, for example, competition for his House seat is strong, he may be in a high risk situa- tion. If his current job is already risky, he may be more likely to seek another office. Schlesinger hypothesized, 38Lester Seligman, "Political Parties and the Re- cruitment of Political Leaders." Political Leadership in Industrialized Societies: Studieggin Comparative Analysis, ed. Lewis J. Edinger (New York: John Wiley & Sons. Inc.. 1967). p. 295. 39Ibid.. p. 299. uoRoland E. Smith. "Political Recruitment and Political Risk in an Oregon Election." (Paper presented to the 1970 American Political Science Association Meeting). P. 5. 26 "We would expect...that those congressmen who did seek to become governor or senator would be men who either came from "“1 So rather than unsafe seats or were sure of advancement. merely examine the competitiveness of the House district in relation to the Senate seat, as suggested earlier, we would also hypothesize that RSTs would be more likely than other Representatives to come from competitive districts. Similarly. if a Representative has low seniority and/ or poor committee assignments, he may feel that a try for the Senate-~even in unfavorable circumstances (i.e. even if he were likely to lose and cost him his House seat)--would not consti- tute "a great loss". Therefore, contrary to our earlier speculation that high seniority and/or good committee assign- ments are important for gaining the resources necessary for a Senate candidacy and hence expecting a disproportionate number of RSTs in these categories, it is possible to assume that poor assignments and low seniority reduce risks and constitute a more attractive situation for a candidate with fewer re- sources. "...When political risk is low, a political career can and will attract more of the weakly motivated and those with more meager resources.“+2 Therefore, it is possible that the hypothesized relationship between importance of committee assignment and the likelihood of progressive ambition may be an inverse one rather than a direct relationship ulAmbitionu and Politics, pp. 17-18. uzLester Seligman, "Opportunity. Risk. Selection and Decision-Making: A Model of Political Recruitments", (Paper presented to the 1970 American Political Science Association Meeting), p. 30. 27 or the curvilinear relationship hypothesized earlier. Not only may an inferior influence position in the House minimize the risk of losing that office. but it may also serve to stimulate ambition for higher office. For instance. Rep. James Scheuer at the time of his candidacy for the Democratic nomination for Mayor of New York City was quoted as saying, "I came into Congress late and I realize I don't have power...But City Hall is something else."“3 So may the Senate be ”something else" with its fewer members, less hierarchical structure, and more evenly distributed power.“u C. Senate Situation The decision of a Representative to become a Senatorial candidate is not merely a function of his own characteristics (or motivations) and of the situation at the "origin" (i.e. the House) which affects his resources, but also of the situ- ation at the "destination" (i.e. the Senate) which provides him with the opportunity. Opportunity is not merely the legal expiration of a Senatorial term after six years. A potential candidate must consider whether his party holds that seat and whether the incumbent plansxto run for re-election. He must consider the likely degree of competition for that seat. He must consider who are the prospective competitors. The latter, of course. is partially influenced by the patterns of office- holding of Senatorial candidates of his party in his state. “3New York Times, June 7, 1969. uuSee Lewis A. Froman. Jr.. The_§ongressional Process: Strategies, Rules, and Procedures (Boston: Little. Brown and Company. 1967). p. 7. 28 All of these aspects of opportunity-~in addition to his moti~ vation and resources-~will determine his decision to become a candidate for the Senate. The motivation or other personal characteristics of candidates who enter certain Senate situations (such as chal- lenging the incumbent Senator of his own party) can be expected to differ from those whose candidacies represent different Senate situations (such as running to fill a vacancy caused by a Senator's death). The House situation (and its resources and risks) can be expected to differ too. And so may still other aspects of the Senate situation-~e.g. incumbents of safe Senate seats may attract more challengers from their own party than occupants of competitive seats. The focus of this portion of the study concerning the effect of the Senate situation upon progressive ambition concentrates upon the HST in one Senate situation as compared to RSTs in other circumstances, rather than comparing him to other members of the House who do not demonstrate progressive ambition for the Senate. It tries, in part. to examine an inference stemming from Seligman and others that high political risk situations.attract different kinds of candidates than low 145 risk situations. Our use of the term here. refers to the chance of defeat. in contrast to Seligman's usage which refers to the cost of defeat. In our current usage, challenging an incumbent Senator in a primary is an extremely risky situation --and we would expect it to attract a different type of RST M “53eligman. "Opportunity.". 29 than would other situations. The classification of Senate situations used here involves two of the aspects noted above-~the party relation- ship of the EST and the incumbent, and the incumbent's re- election plans. The RST is either of the same party as the incumbent Senator, or he is not: the incumbent is either running for re-election, or he is not. Combining these two variables. four different Senate situations are derived: I) Party Challenge-~The EST and the incumbent are in the same party. and the incumbent is running for re-election. II) Party Succession--The RST and the incumbent are in the same party, but the incumbent is not running for re-election. III) Opposition Party-—The EST and the incumbent are in different parties, and the incumbent is running for re-election. IV) Open-ended--The RST and the incumbent are in different parties, but the incumbent is not running for re-election. The assumption made here is that RSTs who compete in Party Challenge-type Senate situations (Group I) are more ambitious than RSTs who seek Senate seats in apparently less risky situations. Therefore, we would expect that these Party Challenger RSTs would be more likely than other RSTs to con- form to the hypotheses noted earlier. Nomination Struggle Our focus to this point has been upon progressive ambition for the Senate and factors which may lead to it-- personal. House, and Senate characteristics, or to state it another way: motivation, resources, and opportunities. But 30 the focus now shifts to part of the struggle to fulfill that ambition-~namely the nature of the struggle for the Senate nomination. Schlesinger's work on ambition theory is not directly concerned with the actual contest for the nomination for the major offices. The opportunity structure for these offices are developed, and the assumption is made "that opportunities arouse expectations and. in turn, give direction to personal ambitions."1+6 The emphasis of ambition theory is "not on hgg one advances in politics but on Egg will want to advance."#7 But the evidence for who wants to advance is drawn from those who have succeeded in gaining the nomination in the past. The opportunity structures--the patterns of previous public offices held by the candidates-~for the Senate emerge from contests for these nominations. The patterns are reinforced or modi- fied by still other contests for the nomination. If 21% of the Senatorial candidates between 1914 and 1958 had Congress as their penultimate office, it is because these 169 Representatives (or ex-Representatives) won their nomination contests.l+8 Some of these primary elections (or convention ballots) may have been unanimous or nearly so, but others were extremely close. Did Representatives from states with certain emerging patterns have more trouble than others? Did Representatives since 1958 have less trouble gaining the “6112213310 anAPontics. p. 15. u7l§$§.. emphasis in the original. “BData was re-calculated from Ambition _gnd Politics, pp. 92 and 950 31 nomination if they were in state parties where Congress was the predominant penultimate office for the Senate nomination, than if they came from states with different patterns? Did Representatives with certain characteristics which are assumed to be advantageous actually have less of a struggle to gain the nomination than other Representatives? An accounting and analysis of the nomination struggles of the RSTs in the two decades from 1950 to 1968 would contri- bute to the answer to these questions. It would also consti- tute a partial test of the assumption that the opportunity structure turns on--and presumably cools off-~expectations and ambitions. The determination of the degree of competition faced by the RSTs in their contests to gain their Senate nomi- nations is essential in the determination of the appropriate legislative strategy for RSTs. This, in turn, is necessary for an assessment of the impact of Senatorial ambition upon the behavior of the RSTs while still in the House. Hence this unit lays part of the groundwork for the last analytical unit of this study. 1 Although some collective statistics are available on the proportion of primaries won by different margins, there is no compendium of accounts of nomination struggles for any series of United States Senate contests. Such a collection would have to refer to newspaper files, books and articles on state politics or on particular candidates, and the occassional case study. Many of these sources have been examined in order to present the brief accounts of each of the nomination con- tests for each of the HST Senate candidacies included in this 32 study. This material is found in the Appendix. Chapter IV is an analysis of some aspects of these struggles. The model described above--and tested below--in part, assumes that certain personal attributes, House behavior patterns, and Senate situation characteristics either consti- tute resources or facilitate their accumulation for use in a Senate contest. These were hypothesized to contribute to progressive ambition. In this portion of the study, these variables are hypothesized to contribute to a successful struggle for the Senate nomination. For instance we specu- lated that being a generalist would be an asset encouraging progressive ambition; now we hypothesize that Representatives with that asset will face less competition in their attempt to gain the Senate nomination. Since the theoretical frame- work here is no different than the one previously outlined, the hypotheses will not be detailed at this point, but rather are noted at the outset of the analysis in Chapter IV. Behavior Change The final focus of this study is upon the impact of Senatorial ambition upon the Representative's behavior while still in the lower house. Schlesinger makes it clear that this is-t part of the development of ambition theory, but that he ignored that problem in his major work Ambition and Politics. ...Ambition theory can be brought to bear on specific overt acts such as the votes of a legislator or the policy proposals of a gover- nor. In this instance ambition theory does not pose a criterion different from the widely held and occasionally tested hypothesis that a legislator responds to his constituency: it differs in that it asserts that the constituency (I) 33 to which the legislator is responding is not always the one from which he has been elected, and that éfiaishfiirfieigfi‘érfiinietih§¥§"h;‘h§§whfsf’33ts t° be The "ambitious politician must act today in terms of the electorate of the office whichluehopes to win tomorrow."50 But alternate strategies of advancement have to be considered. But Schlesinger apologizes, "Regretfully, in this study I shall not do much with the strategies of advancement...Never- theless, it is through the interaction of opportunities and strategies that ambition theory can hope to explain the spe- cifics of legislative or other political behavior."51 Schlesinger notes that ambitions change over time, and "A legislator's votes...are therefore a consequence of current rather than possible future ambitions."52 The examination of the impact of an RST's Senatorial ambition upon his legislative behavior in the House, therefore. can reasonably be limited to his ultimate term in the House. One way of judging the impact is to determine whether his behavior in that term differed from his earlier behavior. But the need to change his behavior would depend upon whether his behavior already conformed to the pattern called for by his strategy of advancement. Two basic strategies were assumed. One was aimed at the party faithful. and would be most appropriate when the candidate faced strong competition for the nomination. This L‘9p. 5. 501bid., p. 6. 51Ibid., p. 16. 52Ibid.. p. 10. 34 so—called Nomination-strategy calls for the HST to be like the other legislators in his state party-~i.e. his party cohesion and liberalism should match the average legislator's of his state party. The second strategy aims at winning votes from the uncommitted and the opposition. and is most appropriate for competitive general election situations. This Election- strategy, therefore, calls for the HST to have a lower party cohesion score than the average state party representative, and to be more moderate. A low level of specialization-- i.e. being a generalist-~18 recommended in each strategy. The prescribed strategy is based upon the level of competition for the nomination as calculated for the analysis of nomination struggles. and the competitiveness of the Senate seat as it was defined for use in Chapters II and III of the study. Our analyses center around two foci--one concerns whether some RSTs are more likely than others to be Ready (i.e. already adhering to their prescribed strategy prior to their ultimate term). and the other examines whether some of the RSTs who are not Ready are more likely to Respond to their new constituency while others are Reluctant to change their behavior. For instance, we would expect RSTs with similar House and Senate situations--e.g. congruent constituencies, or similar levels of competitiveness-~would be more likely to be Ready than other RSTs. Similarly we would expect RSTs from state parties with a favorable opportunity structure-- i.e. Congress is a predominant penultimate office for Senators --to be more likely to be Ready than other RSTs. 35 Our expectations concerning which RSTs would be more likely to change their inappropriate behavior occmasionally stem from contradictory assumptions. For instance, some have viewed lawyers as more likely to have experience in clientele relationships in which they represent viewpoints other than their own and therefore could be expected to Respond to their future constituency. Others do not expect different behavior from lawyers than non-lawyers, and therefore would not hypothe- size different proportions of Responders and Reluctants among these occupational groups. Similarly young Senatorial candidates may be viewed as ambitious and less settled in their patterns and therefore more likely than older RSTs to be Responders. On the other hand older candidates may see their attempt as their only chance to gain the Senate seat (while the younger candidate may still be a viable candidate in the future even if he were to lose), and therefore be more likely to adapt their behavior. An RST with many years of public office experience may be considered a professional who is prepared to do whatever is necessary to gain the office he seeks. and therefore will be more likely to be a Responder than the "amateur" with little previous political experience. On the other hand a man who tries for the Senate without a long record of public office experience may be considered ambitious, and therefore more likely to adapt his behavior to his new needs. These contradictory hypotheses are tested in the final portion of this paper. 36 Study's Population Before presenting any analysis, it is necessary to define in more detail the nature of the population being studied. The time-frame for this study is 1950 to 1968. The group to be studied consists of all members of the House of Representatives who were candidates for the Senatorial nomi- nation. We are not examining Senatorial candidates whose last public office was the House, but rather those who were candi- dates while still in the lower chamber. In states where the Senatorial nomination is decided by primary. there are few problems about identifying those eligible for this study. But in the states that nominate Senatorial candidates in party conventions, challenge-primaries (or convention-primary situa- tions) or situations requiring a special nomination process, there may be some room for dispute of the candidate designation. Therefore special mention must be made of the cases in which the Representative is considered to have run for the Senate nomination, even though he did not risk his House seat. Four Representatives are considered to have been serious contenders for the Senate nomination although failing to secure the nomination in their party's nomination conven- tion. Rep. Andrew Jacobs (Democrat) tried in Indiana in 1950; Rep. Frank Kowalski lost to ex-Representative and HEW Secre- tary Abe Ribicoff in the Democratic convention in Connecticut in 1962: Rep. Samuel Stratton was a serious contender until he got blitzed in the Democratic convention in New York in 1964 by Robert Kennedy's late but successful drive; Rep. Otis Pike of New York was a contender for the Senate 37 nomination in the 1968 Democratic convention, but did not enter the primary which took place subsequently. Jacobs, Stratton, and Pike were able to continue as candidates for re—election to the House, and the latter two were re-elected. Kowalski. who could have run for the House that year as this nomination was purposely not set until after the Senate con- test, refused to do so. When Senator Francis Case died on June 22. 1962, the task of naming the Republican candidate to succeed him fell to the South Dakota Republican State Central Committee. They met on July 9. 1962, and it took them 20 ballots to make their decision. Rep. Ben Reifel was one of six contenders for the nomination. He ran fourth on the first ballot, and was still fourth on the nineteenth ballot, although his position had varied some in the intermediate ballots; he did not receive any votes on the final ballot which saw Lieutenant Governor Joseph Bottum chosen as nominee. Reifel maintained his House seat and was re-elected to the 88th through 9lst Congresses. Another unusual contest occurred when an incumbent resigned in the middle of his term. When Price Daniel re- signed in January 1957 to become Governor, a successor was chosen in an election held in April without any party desig- nations. There were 22 candidates, but three of them received 92% of the vote. The three candidates were Rep. Martin Dies and Ralph Yarborough. both Democrats, and Republican Thad Hutcheson. Yarborough received approximately 38% of the vote. about 75,000 votes more than Dies. Dies retained his House seat. but did not seek re-election in 1958. AV ‘1 ‘J N 38 Another Representative who did not sacrifice his House seat when he made a formal bid to gain the Senate nomination, was John Dent, Democrat of Pennsylvania. He unsuccessfully challenged Sen. Joseph Clark in the 1968 Democratic Senatorial primary while at the same time running for re-nomination to the House of Representatives. He won his House nomination contest and was returned to the 9lst Congress in the November 1968 election. These seven men are considered as having tried for the Senate nomination and are included in the analyses. There were 78 additional Representatives who tried for the Senate nomination while still members of the House. The names of these 85 men, the year of their attempt. and their level of success are noted in the table below.53 53The short-hand expressions used in the code for level of success in that table will also be used throughout the text of the study. RS! stands for Representatives who were success- ful in their attempt to move to the Senate while still a member of the House; RSL stands for Representatives who lost the Sena- torial electioH?IRS combines these two groups; T signifies those Representatives WEB failed to win their bid for the Senatorial :0mination; and RS2 represents all of these sub-groups com- ined. 39 TABLE 1 LIST OF REPRESENTATIVES WHO TRIED FOR THE SENATE 1950-1968 (Code: RSW-Representative was elected Senator; RSL-Repre- sentative was defeated in Senatorial election; T-Representa- tive failed to win Senatorial nomination) YEAR NAME PARTY STATE LEVEL CF SUCCESS 1950 Carroll Dem. Colorado RSL Case Rep. South Dakota RSW Douglas Dem. California RSL Jacobs Dem. Indiana T Kunkel Rep. Pennsylvania T Monroney Dem. Oklahoma RSW Nixon Rep. California RSW Smathers Dem. Florida RSW 1952 Aandahl Rep. North Dakota T Beall Rep. Maryland RSW Beckworth Dem. Texas T Gore Dem. Tennessee RSW Granger Dem. Utah RSL Jackson Dem. Washington RSW Kennedy Dem. Massachusetts RSW Mansfield Dem. Montana RSW McKinnon Dem. California T Potter Rep. Michigan RSW Ribicoff Dem. Connecticut RSL Sasscer Dem. Maryland T 1954 Battle Dem. Alabama T Bender Rep. Ohio RSW Cotton Rep. New Hampshire RSW Curtis Rep. Nebraska RSW D'Ewart Rep. Montana RSL Harrison Rep. Wyoming RSL Howell Dem. New Jersey RSL Hruska Rep. Nebraska RSW Martin Rep. Iowa RSW Sutton Dem. Tennessee T Warburton Rep. Delaware RSL Yorty Dem. California RSL 1956 Davis Rep. Wisconsin T Dodd Dem. Connecticut RSL Young Rep. Nevada RSL 1957 Dies Dem. Texas T* YEAR 1958 1960 1962 1964 1966 1968 NAME By rd Engle Kean Keating McCarthy Prouty Scott Anderson Bentley Burdick McGovern Metcalf Thomson Bass Brewster Curtis Dominick Durno Inouye King Kowalski Merrow Pfost Reifel Seely-Brown Van Zandt Yates Bass Bruce Gill Lloyd McIntire Montoya Stratton Taft Duncan Ellsworth Roncalio Walker Curtis Dent Dole Gurney Hansen Mathias Pike TABLE PARTY Dem. Dem. Rep. Rep. Dem. Rep. Rep. Dem. Rep. Dem. Dem. Dem. Rep. Rep. Dem. Rep. Rep. Rep. Dem. Dem. Dem. Rep. Dem. Rep. Rep. Rep. Dem. Dem. Rep. Dem. Rep. Rep. Dem. Dem. Rep. Dem. Rep. Dem. Rep. Rep. Dem. Rep. Rep. Rep. Rep. Dem. 4O l-CONTINUED STATE West Virginia California New Jersey New York Minnesota Vermont Pennsylvania Montana Michigan North Dakota South Dakota Montana Wyoming New Hampshire Maryland Massachusetts Colorado Oregon Hawaii Utah Connecticut New Hampshire Idaho South Dakota Connecticut Pennsylvania Illinois Tennessee Indiana Hawaii Utah Maine New Mexico New York Ohio Oregon Kansas Wyoming Mississippi Missouri Pennsylvania Kansas Florida Idaho Maryland New York LEVEL or SUCCESS RSW RSW RSL RSW RSW RSW RSW T RSL RSW RSL RSW RSW RSL RSW T RSW T RSW RSL T T RSL T RSL RSL RSL RSW T RSL T RSL RSW T RSL RSL T RSL RSL RSL T RSW RSW RSL RSW T 41 TABLE l-CONTINUED YEAR NAME PARTY STATE LEVEL_OP SUCCESS Resnick Dem. New York T Selden Dem. Alabama T Schweiker Rep. Pennsylvania RSW *Dies was one of 22 candidates running without party desig- nation in a winner-take-all-election—-i.e. the candidate with the highest vote was to be elected. Dies ran second. To summarize the scope of this dissertation: l) we will examine three sets of factors related to progressive ambition for the Senate by members of the House of Representa- tives; these revolve around agent or personal characteristics, the context at the origin or House, and the situation at the destination or Senate; 2) we will examine the nature of the struggle for the Senate nomination; and 3) we will examine the impact of Senatorial ambition upon the Representative's behavior while still in the House. any 42 CHAPTER II AGENT AND ORIGIN CHARACTERISTICS AND SENATORIAL AMBITION Introduction The major focus of this part of the study is to deter- mine whether Representatives who manifest their progressive ambition for the Senate by seeking the nomination while still a member of the House differ from their House colleagues in other ways. The resulting analysis should contribute to our knowledge of political ambition. The model that underlies the analysis in this chapter focuses upon the factors that may lead to the Representative's decision to seek the Senate seat. The model--described more fully in the previous chapter--states that the decision to seek "promotion" from the House to the Senate is a function of three sets of variables. One set revolves around the agent--i.e. the individual's background and personal characteristics, such as his age, his political career. and occupation. The second set of factors concerns his situation at his origin--i.e. the House of Repre- sentatives; it includes such factors as the size and competi- tiveness of his House district. his position in the House as seen in his seniority and committee assignments, and his behavior in the House as seen in his specialization and aspects of his roll call behavior such as party cohesion and liberal- 43 conservatism. The third set of factors contributing to his decision to become a Senatorial candidate revolve around the situation at his destination-~i.e. the Senate: it includes attention to such variables as the party and re-election plans of the incumbent. the competitiveness of the Senate seat, and the opportunity afforded by a previous pattern of Congressmen running for the Senate. Agent Characteristics A. Age One of the personal characteristics examined was age. One line of speculation suggested that the HST. marked by greater ambition than other Representatives, would have started his political career at an earlier age. Therefore one would hypothesize that RSTs gained their first public office at an earlier age than other Representatives; similarly we could expect RSTs to have entered the House at an earlier age too. But an alternate hypothesis, also suggested by Schlesinger, was that the later starters used the advantages gained before entering public office to aid their rapid pro- motion, and that they gained major nominations no later than early starters. An examination of these hypotheses for the RSTs shows little difference between them and other Representatives. The data for other Representatives was based upon a control group that was randomly selected--using a table of random numbers-- from the House membership of the 86th Congress, which was elected in 1958, close to the midpoint of the time period covered by this study (1950-1968). This group of 85 equaled -.~c Iv A: l - ‘ \‘w .v «I! 44 the number of RSTs, and was constructed so as to exclude any Representative who became an RST during the period of the study. This randomly selected group was 67% Democratic com~ pared to 65% for the 86th Congress. This similarity led to the conclusion that this group was representative of the 86th Congress and therefore acceptable as a control group for the RSTs, even though Democrats constituted only 51% of the latter group. The median age for the HST entering public office was 31 which was only two years ahead of the median for the con- trol group. A similar two year difference existed for the age at entry into the House. Although there are some differences in the distributions indicating earlier starters among the RSTs a-e.g. 41% of the RSTs gained their first public office before they were 30. compared to only 33% of the control group, and 39% of the RSTs were in the House before they were 40 compared to 33% of the control group-«these differences are not great enough to alter the conclusion that RSTs do not start their public office career or House career earlier than other Representatives.5u See Table 2. 548cc Walter Oleszek. "Age and Political Careers," prlic Opinion Quarterly XXIII (Spring 1969) Pp. 100—2, and Joseph Schlesinger. "Political Careers and Party Leadership," Pelitical Leadership in Industrialized Socigpges: Studies ingomparative Analysis, ed. Lewis J. Edinger (New York: John Wiley and Sons, Inc.. 1967). pp. 266-293. for other data on age at entry into House office. 45 TABLE 2 Age at Entry into First Public Office and House of Repre- sentatives, RSTs and Control Group (86th Congress) FIRST PUBLIQgQPFICE HOUSE OF REPRESENTATIVES ggg Egg? CONTROL pg; CONTROL 20-24 8 2 - - 25-29 26 26 2 2 30-34 19 29 13 11 35-39 11 6 18 15 40-44 13 9 22 20 45-49 4 10 l9 19 50-54 2 2 6 10 55-59 - l 3 7 60-64 1 - 2 l *Reifel is excluded as his age at entry into public office is unknown. There are some differences within the HST group for first attaining public office depending upon the office. For example. the median age for attaining first public office for the RSTs whose first office was law enforcement was 28, but for local elective office it was 35. The highest. of course, was for Congress (median age at entry was 40) which does have a constitutional minimum age requirement of 25. The medians for RSTs by first public office are as follows: Law enforce- ment: 28, State legislature: 29, Administration: 29, Local elective: 35, Congress: 40. This compares with the control group distribution as follows: Law enforcement: 28, State legislature: 32. Administration: 32. Local elective: 32, Congress: 43. See Table 3. It is interesting to note that the median age for entering the House for RSTs who started there was only one year more than for RSTs beginning their career elsewhere (41 and 40 respect1V81Y). 46 TABLE 3 Median Age at Entry into First Public Office, RSTs and Control Group MEDIAN AGE FIRST PUBLIC OFFICE RST CONTROL Law Enforcement 28 28 State Legislature 29 32 Administration 29* 32 Local Elective 35 32 Congressional 4O 43 Total 31 33 *Reifel is excluded as his age at entry into public office is unknown. Nor were RSTs more likely to seek the Senate seat at an earlier age than others. The median age of all RSTs at the time of their attempt for the Senate seat was 50, which was only a year older than the age of all non-incumbent Senatorial can- didates in a recent year (1964); the median age of Representa- tives who had gained their party's Senatorial nomination was 49. B. Occupation The over-representation of lawyers in legislatures has been documented and discussed frequently. Some of the reasons for the preponderance of lawyers in legislatures would also be relevant to an expectation of their over-representation among the politically ambitious RSTs also. For example, their ability to return to their occupation reduces the risk of defeat and may spur RST status. But an examination of the occupations of RSTs with those listed by angressionai_Quarter y for Representatives in #7 the 84th to 90th Congresses indicates virtually no difference. Fifty-four percent (54%) of the RSTs were lawyers compared to an average of 56% of all House members. The distribution for other occupations also was virtually the same for the two groups of Representatives--with one exception: ll% of all Representatives had teaching listed as one of their occupations, but 17% of the RSTs had been in that profession. See Table u. TABLE a Occupations of RSTs and House of Representatives AVERAGE FOB REPRESENTATIVES OCCUPATION Egg 89th:20th CCNGRESSES* Law 54%** 56% Business 29% 32% Teaching 17% 11% Agriculture l4% 11% Journalism 7% 8% Other Professions 2% 3% *Data from various issues of Congressional Qparterly. **Totals add to more than lOOifibecause some Representatives had more than one occupation. The findings here that lawyers are not more likely than other Representatives to have ambition for political promotion to the Senate supports the general findings of Eulau and Sprague. They conclude, "Lawyer-politicians do not differ much from politicians who are not lawyers."55 Their sugges- tion of convergence of the legal and political profession results in a pursuit of variables other than occupation-- other agent variables, as well as House and Senate situation ones-~to explain the phenomenon of political ambition. 5502, cit.. p. 84. 48 C. Career One of the personal (or agent) characteristics that was assumed to be related to the seeking of Senatorial office by members of the House of Representatives, was the nature of their previous public office career. Schlesinger concluded that those running for higher office had more complex careers. In other words. it is reasonable to hypothesize that Represen— tatives seeking a Senate nomination will have a more complex career than other Representatives. Similarly he found winners to have had a more complex career than losers. Career complexity can be calculated in several ways. One simply is the total number of public offices held; a Representative who has held only one public office prior to being elected to the House may be said to have a simpler career than someone who has held two or three offices. Another way to measure career complexity is to note the different kinds of public office experience a Representative may have had. A Representative who has had only manifest experience-~i.e. prior legislative eXperience--may be said to have a simpler career than another Representative who has had such non~manifest ex- perience as law enforcement or administrative positions. Analysis of the career data for the 85 RSTs and the randomly selected control group chosen from the 86th Congress indicates very little difference in their public office experi- ence profiles. For example, almost half of each group had prior experience in the state legislature (“1% of the RSTs and h2% of the control group Representatives). Approximately one-fifth of each group had administrative experience (19% for #9 each group), and approximately one-tenth had local elective experience (9% for the RSTs and 13% for the control group).56 Approximately one-third had law enforcement experience-~29% of the RSTs and 35% of the control group. It should be noted that each group had virtually identical proportions of lawyers (5b% of the RSTs and 56% of the control group). RSTs, then, are not more likely than other Representatives to have had a more complex career as marked by such non-manifest office experience as law enforcement or administration.57 See Table 5. TABLE 5 Prior Office Experience of RSTs and Control Group 86th COM}. PRIOR OFFICE RSW RSL T RST CONTROL EXEEBIENCE (N=§2) (N: 28) (N=6l) (N: 2h) (N: 85) (1h 85) State Legislature 14 u2%*io 36% 2a 39% 11 46% 35 41% 36 42% Law Enforcement 12 36 9 32 21 34 h 1? 25 29 30 35 Administration 7 21 6 21 13 21 3 13 16 19 16 19 Local Elective 4 12 2 7 6 10 2 8 8 9 ll 13 Statewide Elective 1 3 - - l 2 2 8 3 4 l 1 *Totals do not add to 100% because Representatives may have had more than one previous public office. Nor have RSTs served in more offices than other Representatives. The 85 RSTs held 193 offices (including the 56The definitions of office experience used here were the ones used by Joseph A. Schlesinger in How They Became Governor (East Lansing: Governmental Research Bureau, Michi- gan State University, 1957). Local elective experience, for instance, is limited to mayor, councilman, alderman, school committeeman, and county commissioner. 57This conclusion remains valid when the comparison is made between the RSTs and the entire House membership in the 85th Congress as reported in Schlesinger, "Political Careers," P. 279. 50 House of Representatives) before they tried for the Senate, an average of 2.3 offices. The control group of 85 men held slightly more offices, 214 for an average of 2.5 positions. Both groups had one-fourth of their members come to Congress without any prior public office experience. The original model led to the expectation of more complex careers. Since a politician's career time is limited, it would have been reasonable to expect that RSTs would begin their public office careers in different offices. But in view of the lack of difference in total office experience between RSTs and other Representatives, it is not surprising to find that there were virtually no differences between them concer- ning their first public office experience. Identical propor- tions--25% started in Congress, and the difference between RSTs and the control group never varied more than three percentage- points (15% of the RSTs started in administrative positions, compared to 12% of the control group). Approximately 30% (31% of the RSTs and 29% of the control group) started in the state legislature--more than any other office; approximately one-fourth (24% of the RSTs and 26% of the control group) begain their career in law enforcement positions. Only 6% (RSTs) and 8% (control group) began in local elective office despite the large number of available positions, while no one began in statewide elected positions. See Table 6. It appears, therefore, that one cannot differentiate RSTs from other Representatives by examining the kind of public office in which they began their career, or the offices in which they served prior to their Rouse membership. 51 NOON mm m@1:..MI NNH 0H Nmm am New NN Nam mm Hoausoo mwmwmcoo zoom Hmm mm “‘I ma HN ON om NHOH Nma Nmm Nma Nmm «N Nam He NOON mm wmnn. ml. mmzu._mn Nos ea Naa m Nam ma NHN e New on Nam N Nam NH Now a mm Amm NOON mmwn. Nma Nam NNN NNN mm m mHmBOH m>HuomHm Hmooq Coaumuumficwap< mmmamcoo unmamouomam 3mg musumamwmmq mumum 3mm aaowo Houeeeo eee mama me eeammo eaaaaa sauna o mdmwumucmmmwamm HH< use .msmatmz muamm mumum .mHmm mo mEMmH mo wswumm m>aum>wmmCOOIHmwman 0H mqm no> .>aomsoo " o Hmamnflq u A .eaa sum> un> mwdwumm HmowwoaomvH 87 Republican RSTs as more liberal: 49% of the median state party Republican terms were L or VL compared to 62% of the RSTs'. But Democratic RSTs are no longer more liberal than their party colleagues: although only 53% of all Democratic terms are L or VL compared to 68% of RST terms, 72% of the median Democratic terms are L or VL. In other words, Democratic RSTs are slightly more conservative than their median state party colleagues. See Table 16. Another way to show this comparison is to rate the score of each RST term in relation to his state party median's score-~i.e. the RST is either More Liberal, the Same, or Less Liberal (or More Conservative) than the state median on the liberal-conservatism continuum. Using this measure, Repub- lican RSTs are still more liberal than their state party median colleagues: 42% of the time the HST and the state party median are in the same category, but in 38% of the terms the HST is More Liberal in 18% of the terms but More Conser- vative in 20% of the terms. See Table 17. TABLE 17 RST Liberalism-Conservatism in Relation to State Party Median for All Terms, 80th - 90th Congresses RST COMPARED TO HIS STATE PARTY MEDIAN IS: DEMOCRATS REPUBLICANS More Liberal 23 18% 54 38% Same 77 62 61 42 Less Liberal 25 20 22 .29. 125 100% 144 100% Rather than compare terms of service, which gives 88 a senior man more weight in these results--for instance, Keating alone contributes 6 of the 57 VL RST terms and 6 of the 54 More Liberal ratings for the Republicans--a single average score for each RST can be used.82 In view of the assumption that an RST's record will influence his decision to seek the Senate nomination, the scores used in computing this average does not include his last term in Congress (and does not precede the 80th Congress)?3 The earlier findings remain--Republican RSTs are more liberal than their state party median colleagues, while the Democra- tic RSTs are barely more conservative (or the same-~depending on the method used) than their party colleagues. Fifty-four percent (54%) of the Republican RSTs were L or VL compared to 42% of the median Republicans from the same states. One- third of the Republicans averaged a higher or More Liberal score than the median, while only 18% were lower or More Conservative (48% agreed on their average rating). For the Democrats, 73% of the RSTs were L or VL compared to 76% of the median Democrats. But equal proportions of RSTs were higher as were lower, suggesting no over-all difference between Democratic RSTs and the median Democrats (in fact 68% agreed on their averaged rating). See Table 18. 82The arithmetic mean of the category "grades" with an equi-distant scale was used: if the mean was equi- distant between two categories, the score of the last term was given double weight. 83The analysis also excludes Dies whose attempt for the Senate nomination occurred in the first session of the 85th Congress: it also excludes l4 RSTs who made the bid for the Senate in their freshman term. 89 TABLE 18 Liberal-Conservative Average Rating for RSTs, 80th Congress to RST's Penultimate Congress DEMOCRATS REPUBLICANS STATE STATE PARTY PARTY AVERAGE RATING RST MEDIAN RST MEDIAN Very Liberal 14 38% 13 35% 11 33% 5 15% Liberal 13 35 15 41 7 21 9 27 Conservative 9 24 9 24 9 27 16 48 Very Conservative _; 3 _; *6 18 _3 9 37 100% 37 100% 33 99% 33 99% RST COMPARED TO HIS STATE PARTY MEDIAN IS: More Liberal 6 16% 11 33% Same 25 68 16 48 Less _6 16 _§ 18_ 37 100% 33 99% The conclusion that Republican RSTs are more liberal than other Republican Representatives, while Democratic RSTs may be considered more conservative than other Demo- cratic Representatives may stem from a number of factors. For instance, some analysts have associated youth with liberalism.84 Others have hypothesized a relationship between competitiveness of one's district and liberal- conservatism. For instance the speculation of Downs, and others, would lead us to believe that legislators from marginal districts must be moderate in order to attract 9O voters from the opposite party (i.e. Republicans will be more liberal, and Democrats more conservative). And of course the thrust of Schlesinger's work is that the nature of the office which the legislator seeks--in this instance, the competitiveness of the Senate seat-~will affect his behavior. A review of the data reported earlier in this chapter indicates that the RST is probably younger in view of his lower seniority (as compared to all members of the Blst to 90th Congresses), is more likely to come from a competitive district (as compared to all elections for Representative in that period), and he is less likely to be seeking a safe Senate seat (at least as compared to the control group in the 86th Congress). These distributions would indicate that the pattern of RST ideology can be ex- plained by the three variables noted: age, House competi- tiveness, and Senate seat competitiveness. But a closer examination of the data indicates that the age of the RST and the competitiveness of his House seat cannot explain why Republican RSTs are more liberal and Democratic RSTs are more conservative than their party col- leagues. For instance, Table 19 indicates that younger RSTs 8“Bug. Angus Campbell, et.al., The American Voter (New York: John Wiley & Sons, 19607: pp. 210- ll. 91 are not more liberal than older RSTs. In fact older Republican RSTs are more liberal than younger Republican RSTs, while younger Democratic RSTs are as likely to be more liberal as to be more conservative than the state party median representative. Similarly, Table 19 indicates that contrary to the hypothesis, RSTs from competitive House districts are not more moderate than RSTs from safe districts. Again the data for Republican RSTs is opposite the hypothesis-~safe Republi- can RSTs are actually more liberal than competitive Republican RSTs--while the data for the Democratic RSTs indicates equal distributions of more liberal and more conservative among safe and competitive RSTs. In contrast to these two failures to link the explanatory variable to the RST pattern, is the data on the Senate seat competitiveness. Republican RSTs seeking a com- petitive Senate seat are more likely to be liberal than those seeking a safe Senate seat as hypothesized, while Democrats seeking a competitive seat are more likely to be conservative than those seeking a safe Senate seat as hypothesized. Although these comparisons are based on small num- bers because most RSTs agreed with their state party median (68% of Democratic RSTs were in the same category as the median, and 48% of the Republican RSTs were), and the dif- ferences may not appear large, they are in the hypothesized direction and therefore suggest a possible relationship between the Senate constituency and the behavior of the RST while he still is in the House. 92 UHQQ Hmhmflflg who: .mwmucm min VAQHUHHHV$ one Roam mmmusmoama Hewenflq mmmA one mafiuomuunsm an nouosaumcoo . mm NOQH e - NNe NNN HH- Nee Nmm Nee NNN N mmmHeeom . . - e N e N e NH NHoH NeH Nee Nam eo- NQOH NNH NNN NHH e co . mH N N e NH m NH N e>HoHee Bee Nee NNN Nme NNN ON NOQH - New NON m HH m m m 0H - N N mm m Hummm cumsom o N . . o o \HeH N NNN N Nme m NNN 00 NQOH NNN New NNN Hoemaoo . . . N e N m N eeHo. 0N NOOH oN NmH New Nmm oo Nee NeH NHN NeH e NH e NN e oN e deem "uoaaumwa mmsom m N . . . N \ooH ON NoH N Nam HH Nmm N 00 NQQH NHH NNN NHH o o m ..T 0 oo NOOH NH NHN e Nmm NHN oo NQQH NNH H Nee N NNH H A Omv eH m e NN m NH m Hon Heedev mazes Nm C . e H Hmooa mMWH deem .BHH exeedH Hedda .BHH :wsmm .omm A oaoz mmoA mums mHmm cmoNHnsaom mHmm oHumaooEmm mmmawc mmwmwwwefiwwmsmm m.Hmm ou mmmuwsoo :uow .msmfitoz auamm mumum ou UonQEoo wmm “Dem mwmwm>w o>wum>aowaoonamamnaq tam .mmmsm>aufiumaaoo m mumsom tam wmosm>fiuwumanu uoNaumNQ mmdom .ow< ma mqmv sowuflmom amaom mmmawsoo mumafiuassmm m.Hmm cu mmmuwcou Show .mHmm pow mwsfiumm mwmam>< o>aum>umNSOO:HmaonNA Use .mmmso>wuwumano uoflwumwo omsom tam mw< mHmm smoflansmmm NOOH O NOOH mm NOOH O NOON mm Hence New N New NH NNN N NHN oN pesto NNO NON w NHmuerHH %Hm>v coauwmom amaom mHmm oflumaooEmQ ON mqmwuwumanu mwmm HuoNaDmHO mmsom A+ ONO eHo AOm amnssv wssow umw< 96 in the Very Liberal quartile if a Democrat and the Very Conservative quartile if a Republican) than other RSTs: 6) unexpectedly-~but similar to Fishel's finding for non- incumbent Congressional candidates--RSTs from competitive districts were more likely to be "ideologues" than those from safe districts. H. Party Cohesion Party Cohesion is the last variable to be examined to determine whether Representatives who seek a Senate seat (RSTs) differ in other ways from their House colleagues. Our model relating Senate candidacy in part to factors related to the Representative's behavior in the House, does not suggest any basis upon which to speculate concerning the party cohesion level of the RST. It is possible to hypothesize that Repre- sentatives with progressive ambition for the Senate would be more likely to be loyal to their party, because they realize they need the support of their party in order to gain this new office. But American political parties are rarely so well organized that they can deny the nomination and election to a candidate who has demonstrated a low level of party loyalty in his previous legislative voting record. It is possible, therefore, to speculate that RSTs who must appeal to a new and larger constituency will be best served by a previous voting record indicating indepen- dence-~i.e. low party cohesion. Men with shifting consti- tuencies due to altered ambitions may look to themselves for guidance on votes rather than to their party. Soule's data, 97 for instance, suggests that state legislators who have progressive ambition are more likely to view themselves as Trustees than Delegates.87 Trustees are less likely to be loyal to their party, Davidson discovered in his study of the House of Representatives.88 The third alternative hypothesis-~that RSTs will Show average party cohesion, neither more nor less than the average Representative of the same state and party--is also logically defensible. The demandeof the statewide consti- tuency may be most adequately represented by the behavior of the average or median state-party legislator. If so, then the party cohesion of the RST would be most likely to match that of the median state-party legislator. In order to test these divergent hypotheses, a methodology using roll call party cohesion scores had to be developed. There are several methodological factors to be considered in any analysis using a comparison of roll call party cohesion scores. One of these is the choice of roll calls. Users of roll call cohesion scores have utilized several different operational definitions of a party roll.call. For example, one could use all roll cells including unanimous ones, or they could use only non-unanimous ones, or those in which a majority (or any other proportion) of one party opposes a similar prOportion of the other party. We have used the data found in Congressional Quarterly (pg), and their definition of Party Unity roll call 87Soule, op, cit,, p. 452. 88Davidson, op, cit., p. 161. 98 is one "on which a majority of voting Democrats opposed a majority of voting Republicans." This definition apparently was unchanged throughout the period of their studies. A second element to be considered is what constitutes a vote. Unfortunately 93 did change this operational defini- tion. From the 80th to 83rd Congresses, they included taking any stand on the issue by voting, pairing, announcing his position, or even answering a QQ poll on the question. But beginning with the 84th Congress they used only yea and nay votes, and excluded the other means of registering a position. Hence the results for the Congresses prior to the 84th and those subsequent to it are not exactly comparable. This study-~as others have-~has assumed their comparability, and no scores were re-computed as a result of their change in definition. A third methodological question concerns the denomi- nator upon which the cohesion score is based. Prior to the 84th Congress the denominator was the number of votes cast by the legislator (using the broad definition of a "vote"). A legislator's party unity score was the number of times he voted with his party divided by the number of his votes with and against his party (times 100)--i.e. (Y/Y&N)(100). But beginning with the 84th Congress the denominator was the total number of party unity votes for which the legislator was eligible. To illustrate: if there were 100 party unity votes and a legislator voted with his party 70 times, and against it 20 times, and was unannounced on 10, then for the 80th to 83rd Congresses, Q3 would calculate his score as 99 78--i.e. (70/90)(100): but from the 84th Congress onward his score would be 70--i.e. (70/100)(100). There is a second reason, therefore, why scores prior to and subsequent to the 84th Congress are not exactly comparable. This change in methodology is especially relevant to this study because candidacy for the Senate would probably result in a great reduction of votes cast because of the necessity of campaigning for the nomination in a statewide primary (or just traveling to get acquainted with the new electorate). In the earlier methodology absences or general pairs did not effect the cohesion score: but in the current 93 methodology, failure to vote would decrease the cohesion score. Therefore, it was necessary to use a consistent metho- dology. The cohesion scores of the RSTs for the 84th Congress onward were re-calculated using the same denominator as used in the 80th to 83rd Congresses.89 For example, in the 90th Congress Rep. Resnick, deeply involved in a campaign to win the Senate nomination in New York, voted with his party on only 63 (or 64) votes out of 171 party unity votes while never opposing a majority of his party. His party unity score was considered to be 100 rather than the 37 reported by QQ. Still another methodological factor must be noted in our use of the roll call cohesion scores reported by 93. The 89For the 83rd Congress 9Q reported a legislator's party unity score only and not how many votes he cast with his party. Therefore the scores in the 83rd Congress could not be converted to the later methodology, but those of the 84th to 90th Congresses could be converted to the earlier methodology. lOO meaning of a numerical cohesion score (e.g. 85) varies from one Congress to another depending, in part. upon the nature of the bills and amendments subjected to roll call votes. For instance, the median cohesion score for the Democrats in the 82nd Congress was 87, but it was only 80 in the next Congress. Therefore the scores for each party in each Con~ gress were categorized into quintiles ranging from highest cohesion (A) to lowest (E), based upon the scores at the 20th, 40th, 60th, and 80th percentiles.90 In other words, except for tie scores at the dividing points, each party in each Congress had one-fifth or 20% of its members in each of the five categories. Therefore a score of 90 by a Lamocrat in the 82nd Congress was only a "C", but it was good enough for an "A" in the 83rd Congress. In effect this operation com- pares an RST's scoreto all of his party colleagues. There are several methods of comparing the roll call party cohesion quintile scores, and these will be discussed as they are used in the presentation of the analyses. Categorizing the scores of all Representatives of the same party in each Congress into quintiles, means that reporting the distribution of scores for RSTs automatically 90This is not totally accurate for the 84th to 90th Congresses. The score used to determine the dividing point between quintiles was actually an artificial one constructed from party unity and party opposition scores at the appro- priate percentile. For example. for the Democrats in the 90th Congress the 20th percentile party unity score was 83 and the party opposition score was 4: this results in a 95 (Y/Y&N)(100) as the artificial score at the 20th percentile required for a score of "A": it is possible that no Demo- crat actually had an 83-4 score. lOl constitutes a comparison to the RSTs' colleagues. Using this method for all terms of the RSTs between the 80th and 90th Congresses,91 the HST appears to resemble his colleagues. The five categories contain between 19% and 21% instead of the 20% expected hypothetically. Some partisan differences appear, however. Democratic RSTs are very unlikely to be in the low- est cohesion category (E) (only 6% of all terms compared to 22-26% for the other categories) while Republican RSTs are most likely to be in this category (31% compared to 16-19% for the other categories). See Table 21. TABLE 21 Party Cohesion Level of Terms of RSTs, and Their State Party Medians, 80th-90th Congresses TOTAL DEMOCRATS REPUBLICANS PARTY STATE STATE STATE COHESION PARTY PARTY PARTY LEVEL RST MEDIAN RST MEDIAN RST MEDIAN A (High) 56 21% 27 10% 32 26% 24 19% 24 17% 3 2% B 52 19 88 33 29 23 50 40 23 16 38 26 C 57 21 83 31 29 23 29 23 28 19 54 38 D 52 19 51 19 27 22 19 15 25 17 32 22 E (Low) 52 19 20 2 8 6 _3 2 44 .31 17 12 269 99% 269 100% 125 100% 125 99% 144 100% 144 100% INDEX: AB 4% 42 4g 59 33 28 -DE -3 -2 -2 -12 - 8 - 4 + 2 +17 +21 + 2 -15 -36 A more meaningful comparison, however, is between the RST and the median Representative of his state and party. Comparing the distribution of terms (See Table 21) for RSTs and their state-party medians, we see that both the Democratic 919g--the source for this data-~did not report party unity scores prior to the 80th Congress. 102 and Republican RSTs are lower in cohesion than their col- leagues. For example, 49% of the Democratic RSTs' terms are in the two highest categories, while 59% of the Democratic median Representatives' terms are: 28% of the Democratic RSTs' terms are in the two lowest categories compared to only 17% of the median Democrats'. Republican RST terms are both more likely to be highly cohesive (33% of the RST terms compared to 28% of their colleagues') and low in cohesion (48% to 34%): but an index constructed by subtracting the proportion in the two lowest levels of cohesion from the proportion in the two highest levels shows the Republican RSTs are lower in party cohesion than the median Republican legislators. This comparison between the party cohesion level of each term for RSTs and that of his state party median can be combined into a single figure. Table 22 indicates that the party cohesion level for Democratic RST terms is more likely to be lower than his state party median than to be higher: the data for Republicans:aresimilar. Thirty-eight percent (38%) of Democratic RST terms are lower compared to 28% which are higher: in 34% of the terms the RST and the state party median are in the same category of party cohesion. For the Republicans, 33% of the RST terms are lower compared to 29% which are higher: there was agreement in 38% of the cases. 103 TABLE 22 RST Party Cohesion in Relation to State Party Median for All Terms, 80th-90th Congresses RST COMPARED TO HIS STATE PARTY MEDIAN IS: DEMOCRATS REPUBLICANS Higher Cohesion 35 28% 42 29% Same Cohesion 42 34 55 38 Lower Cohesion 48 _38 42 33 Total 125 100% 144 100% But the use of terms as the base of comparison of RSTs and the median representatives of their state and party, gives senior men more weight and so may not reflect an accurate picture of the relative level of party cohesion of the RST. Table 23 reflects the data using a single average score summarizing an RST's cohesion scores between the 80th Congress and his penultimate term in the House.92 The average party cohesion scores for 54% of the Democratic RSTs were in the two highest categories (A and B) compared to only 30% in the lowest two categories (actually no Democratic RST averaged in the lowest quintile). But the average scores for the median state party representatives were even higher; 65% were in the two highest while only 8% were in the two lowest. Democratic RSTs, therefore, can be considered less loyal to their party than the average Democratic Congressman. The same is true for the Republican RST--he shows a slightly lower level of party cohesion than the median Republican 92As in the similar analysis of liberal-conservatism, an arithmetic mean of the categories using an equi-distant scale was constructed, with the last term given double-weight when the average was equi-distant between two categories. 104 representative. Thirty-six percent (36% of the Republican RSTs were in the two highest categories, while 42% were in the two lowest categories for an "index" of -6. Of the median Republican legislators, 21% were in the two highest cohesion categories, and 24% in the two lowest, for an index or -30 TABLE 23 Party Cohesion Average Rating for RSTs, 80th Congress to RST's Penultimate Congress DEMOCRATS REPUBLICANS PARTY STATE STATE COHESION PARTY PARTY LEVEL RST MEDIAN RST MEDIAN A (High) 9 24% 7 19% 6 18% - -% B 11 30 17 46 6 18 7 21 C 6 16 10 27 7 21 18 55 D 11 30 3 8 10 30 7 21 E (LOW) _: - __:_ - i 12 ___]_._ 2 37 100% 37 100% 33 99% 33 100% INDEX: AB 54 65 36 21 :QE ' 0 :_§ ““2 :33 +2 + 7 :_3 - 3 The data in Tables 21-23 demonstrate in different ways that the RST-~whether he is a Democrat or a Republican --shows a lower level of party cohesion than the median rep- resentative from the same state and party. The net differences between higher cohesion and lower cohesion are smaller for the Republicans than for the Democrats, regardless of the method of presenting the data. But when we use a single figure to represent the average cohesion level of the RST compared to average cohesion of the median representative of the same state and party as 105 the RST (as constructed by averaging these comparisons--i.e. Higher, Same, Lower-~for each term from the 80th Congress to the RST's penultimate term), this finding is altered partially. Democratic RSTs are still seen as being less loyal to their party than the average Representative of the same state and party. But Republican RSTs are now viewed as more cohesive than their party colleagues. Table 24 indicates that 33% of the Republican RSTs average Higher compared to 24% who average Lower, while 42% are the Same as the median state party rep- resentative's average. (For the Democrats the comparable data is 24% Higher, but 38% Lower, with 38% Same.)93 The proportion of RSTs who are the Same as their state party median is as large or larger than the proportion who are either Higher or Lower than their median colleague. Nevertheless, in each party the proportion of RSTs who differ from their state party colleagues is greater than the propor- tion who are the same. Therefore it is reasonable to measure whether RSTs are more or less cohesive than their party col- leagues through construction of an index based on the propor- tion of Higher RSTs minus the proportion of Lower RSTs; a positive number indicates more party cohesion and a negative number indicates less party loyalty. The literature on party cohesion contains contradic- tory hypotheses concerning the effect of competitive districts upon the level of party cohesion. Some have suggested that 93The findings are basically the same even if the average of the RST's terms were compared to the average of his state party's median; in fact Republican RSTs are even slightly more likely to average Higher--39% compared to 33% (and a Higher-Lower index of +12 instead of +9). 106 mm+ u o o o - o o u - \oon aaa amm aom om aooa aem \es mamamaom mH+ \ o H N m m m \ m q \NN n whoa a u m - m s m as n ma ma m \qq N \wm 0 NH \ooH ma m N N q m>fluwumano we: Nam Non qu NwH Om+ NOQH New Nom Nom m mmmm m an e m N on N “mam mumsmm nm+ Nam N H n w - u Nmm Nae NNN N d H \GH H \HN m HH \mm m m d N Q>HuflquEOU so: NooH NNN Now Nmm «an Nooa Nam Nom NmN m 8N a ma 0 mm as OH a mm m .umaa mmsom xmch HmDOH .300 mEmm .Soo xmch HmDOH .Lou mEmm .Loo wm3oq umnmam HmBOq umsmwm Noon mm Nooa mm. Hmuoa IMMI. Ml mm «a soammnou nmzoq Nd «H mm «H soammsoo mamm Nmm Ha New m cowmmaoo wmnwflm ”ma swaps: muumm mumum mam ou @mumano Hmm mcmowabsmwm mumwooEmQ mmmsm>wuflumano ummm mumsmm paw mmmcm>wuwumaaoo uofluumaa mmsom tam .Ammmuwaoo mumafluadcmm m.Hmm on mmmwwcou auowv msmwpmz sawmm mumum on pmwmmfioo mHmm wow mwcHumm cowwmgoo muwmm «N mdmhmm mo mahmu HmDOH xmecH Hence .no emuumm .no wxweaa Hence .no emuumm .no w .ommH uxm auwmm .omvH uwm muwmm mHmm :moNabsmum mHmm owumwooamo aoNumucmeo muwmm was .ucmEdwNmm< mmuuNEEoo .muNHONcmm 0N mdmmH 114 also, for the Republican RSTs on three of the four measures used; but when a single average was used to compare the RST to the median legislator in his state party, it indicated more Republican RSTs in the Higher Cohesion category than the Lower Cohesion category. For both parties, RSTs from safe House seats were less cohesive than those from competitive districts; apparently safe RSTs were freer to desert their party and to develop an appeal to a wider constituency. Demo- T crats seeking a safe Senate seat were more likely to be High cohesion than other Democratic RSTs, but Republicans seeking safe seats were least loyal to their party. At first it was suspected that this was due to the finding that RSTs seeking competitive Senate seats were more moderate than those seek- ing safe seats. But when a measure was devised to isolate party loyalty from the ideological dimension, this basic pat- tern still persisted~~i.e. Democratic RSTs seeking safe Senate seats still were more party-oriented than other Demo- cratic RSTs, while Republican RSTs trying for a safe Senate seat were less party-oriented than other Republican RSTs. tsing this party-orientation data, it was noted that Repub- licans were more party-oriented than the Democrats. Repub- lican RSTs serving on the lowest-rated committees, and those with the lowest seniority, were most likely to be party- oriented. This was not surprising as Matthews' data on the party effort scores of Senators had also indicated that party effort declined with seniority. But the data for Democratic RSTs were contrary; those with the least service were the least party-oriented. and those with the lowest-rated 115 committee assignments were least party-oriented. Conclusion Our model posited the decision by a Representative to become a Senatorial candidate as a function of three sets of variables~~agent or personal characteristics, origin or factors associated with service in the House of Representa- tives, and the situation surrounding the destination or Senate seat. This led to an investigation of whether RSTs can be distinguished from other Representatives by their personal background, or House situation. The data on personal characteristics indicated that there were few differences between RSTs and other Represen- tatives. But the data on House situation indicated several differences-~although some were not statistically significant. RSTs were more likely to come from proportionally larger dis- tricts (i.e. At-Large seats or districts that represented 20% or more of the state's population) and larger cities than other Representatives. They were more likely to come from competitive districts than other Representatives, as was hypothesized. But contrary to expectations derived from Schlesinger's work, RSTs were more likely than other Repre~ sentatives to come from organizational tension situations created by different degrees of House and Senate competition. RSTs had less seniority than other members of the House, suggesting minimal risk may be a more important stimulus to Progressive ambition than high resources. But some limita- tions to this thesis can be seen in the fact that RSWs had 116 more seniority than other RSTs; it was noted, too, that those with very high seniority and those with very low seniority had lower success rates than those with moderate service. A more direct challenge to the minimal risk thesis was the data indicating that RSTs had better committee assignments than other Representatives. But few chairmen or ranking minority members sought Senatorial office, and there was some indica- tion that RSTs with the best committee assignments were most likely to seek safe Senate seats~~suggesting that they were reluctant to risk their positions in uncertain situations. But the analysis of legislative behavior~~as distin- guished from these other aspects of the House situation~~was less clear concerning differences between RSTs and other Representatives. There was no comparable data on Speciali- zation for non-RSTs, but comparisons between the different sub-groups of RSTs (i.e. RSWs, RSLs, and Ts) suggested no clear relationship between specialization and progressive ambition for the Senate. The data on liberal-conservatism indicated RSTs were more likely than not to be in the same quartile as the median state party legislator; all RSTs were in the same quartile as their median colleague 51% of all terms, but only 42% of the Republican RST terms showed agreement between the RST and his median colleague. Similarly although 59% of all RSTs averaged (for the 80th Congress to their penultimate term in the House) in the Same category as their median, only 48% of all Republican RSTs did. So although Democratic RSTs were more likely to agree than disagree with their colleagues, 11? Republican RSTs were more likely to disagree than agree, however more were the Same than either More or Less Liberal. The data on liberal-conservatism also showed Repub~ lican RSTs to be more likely to be liberal than conservative, while Democratic RSTs compared to their median were more con- servative if there was any net difference. RSTs seeking competitive Senate seats were more moderate than those seeking safe seats. There was no consistent relationship for both parties between the RST's age or the competition of his House seat and the level of his liberal-conservatism relative to his state party median. But it was found that RSTs from com- petitive districts were more likely to take an ideologically polar position than safe RSTs: similarly older Democrats and younger Republicans were most likely to take the polar posi- tion. The roll call data on party cohesion wereclasslfied into quintiles instead of quartiles. One result was a lower proportion of RSTs who were in the same category as their state party median. In only 34% of the terms did the Demo- cratic RST agree with his median, while 38% were Lower; 38% of the Republican RSTs agreed, which was more than in either the Higher (29%) or Lower (33%) categories. Similarly 38% of the Democratic RSTs averaged (for the 80th Congress to their penultimate term in the House) in the Same category as their median, while an identical proportion were Lower; for the Republican RSTs more were in the Same category (42%) than in either the Higher (33%) or Lower (24%) categories. Other findings on party cohesion indicated that RSTs 118 from safe seats were less cohesive relative to their median colleague than RSTs from competitive seats. Democrats trying for a safe Senate seat were more cohesive than those trying for competitive seats, but it was reverse for the Republicans ~~safe Senate seats and lower cohesion were associated for them. This was true even after party orientation was isolated from ideological considerations. Using the data on party- orientation indicated that Republican RSTs were more loyal to their party than Democratic RSTs (relative to their state party medians). But it also showed that half of the RSTs were in the relative category expected based upon their cohesion and ideological position. It also indicated that for Repub- lican RSTs low seniority and poor committee assignments were associated with higher party-orientation, while for the Demo- crats low seniority and poor committees were associated with lower party~orientation. Although there are some differences between RSTs and other Representatives on the basis of factors associated with the situation at the origin or House of Representatives (less when it comes to roll call behavior than other aspects), the impact of the Senate situation upon the Representative's decision to seek a Senate seat must be examined. 119 CHAPTER III DESTINATION SITUATION AND SENATORIAL AMBITION Purpose The focus of this chapter is upon the third set of factors assumed to affect a Representative's decision to become a Senatorial candidate~~the opportunity afforded by the situation at the destination office (the Senate). The legal opportunity of a vacancy~~due either to the ex- piration of the term of office or the death of the office- holder~~is essential, but other aspects of the opportunity situation must be examined too. Which party holds the Senate seat, and the re~election plans of the incumbent are two important Questions of concern to the potential candidate. He must also consider the prOSpects of election-~i.e. the competitiveness of the Senate seat~~as well as anticipate the amount of competition from prospective opponents for the nomination. These potential candidates (RSTs and opponents) may be influenced by another aspect of opportunity-~the career patterns found in their state party. This may be due to the candidate's internalization of certain expecta- tions about Senatorial careers stemming from past patterns. Or the pattern may serve to put potential candidates into the forefront of speculation~~a position of noteriety that 120 may be considered one of his resources or advantages which may enable him to attract other needed resources such as financial and organizational support. The Senate situations will vary from case to case. In some situations, the RST will seek the Senate nomination in a very hospitable environment, while in other situations the RST will face a hostile environment. We assume that the RSTs deciding to run in the latter situation will differ from RSTs in the safer situation. This assumption stems from Seligman98 who suggests that low risk situations attract different kinds of candidates than high risk situ- ations. The focus in this chapter is upon one group of RSTs compared to other groups of RSTs rather than to other House members without progressive ambition for the Senate. The classification of Senate situations used here is based upon the first two elements noted above~~which party holds the Senate seat, and the incumbent's re~elec- tion plans. These can be combined into a single matrix. The RST is either of the same party as the Senate incumbent, or he is not: the incumbent Senator is either running for re~election, or he is not. A further word on the operational definitions of incumbency and candidacy is necessary, however. Operation- ally, "incumbency" refers to the occupant of the Senate seat, regardless of the nature of his selection~~i.e. whether he was elected to it, or appointed to fill a vacancy. The 98“Opportunity", opacit. 121 choice of this definition is based upon several factors. If the appointee is running for re~election, then the RST faces the situation of comparing his Congressional voting record to that of the incumbent. If the incumbent is not running for re~election, it would make little difference to the HST whether this was due to the death of an elected Senator or the retirement of an appointed one. This is not to say an appointed incumbent seeking re~election has all of the advantages that an elected one has. Not having faced the electorate of the Senate consti~ tuency may be an important difference. The length of the term, however, is not related to the election/appointment issue. On occassion appointed Senators have had more seniority in the Senate when they faced re~election than have duly elected Senators. For example Sen. Ed Mechem of New Mexico was appointed to the Senate on November 30, 1962 and ran for re~election in November 1964; Sen. Ted Kennedy was elected to fill an incompleted term in November 1962, was not sworn until January 3, 1963, and also sought re~election in November 1964. In addition, it should be noted. we make no other references to length of service in our classification ~~an RST opposing a six-year Senator is not classified dif- ferently than one facing a three-term veteran. The candidacy status of an incumbent (i.e. whether the Senator is “running" or "not running") refers to the situation at the time of the RST's decision to become a candidate. If an RST resolves to challenge the incumbent in his own party's primary, we consider him very ambitious 122 to be ready to run in such a risky situation. This evalua- tion of his ambition should not be changed if the Senator subsequently withdraws from the race. For instance Rep. Helen Gahagen Douglas (Democrat) decided to challenge Sen. Sheridan Downey (Democrat) for the Senatorial nomination in California in 1950; we would expect her to share certain characteristics with other RSTs who decided to run against ! incumbents. There is no reason to change our expectations __.-—- of her characteristics or pre-candidacy behavior simply be- cause Sen. Downey subsequently withdrew from the race. If a Senator announces his retirement after the HST has entered the race for the Senatorial nomination, then the situation is classified as "incumbent running", even if he is not a candidate on Election Day or even on the Primary Election ballot. In the example cited above, Rep. Douglas is classified with other RSTs who were in the same party as the incumbent who ran for re~election. Rep. Richard Nixon (Republican) who also had announced his candidacy for Sen. Downey's seat is considered to be in the same category as other RSTs who decided to (and did) run against the incumbent Senator of the opposite party. Similarly, an RST's category does not change because the incumbent is eliminated in his own party's primary. Rep. Thomas Curtis (Republican) who decided to challenge Sen. Edward Long (Democrat) only to find himself running against a different opponent when Long lost his primary battle remains in the group which chose to oppose incumbents. One unusual case was that of Rep. Clarence Young 123 (Republican of Nevada). He had announced that he was not a Senatorial candidate because Democratic incumbent Sen. Alan Bible was expected to be a candidate. In December Bible said he would retire; Young then became a candidate. But in May Bible decided to run after all. This situation was classified as incumbent not running. Party similarity or difference, of course, refers to the designation of the HST and the incumbent (regardless of how the latter attained office). If the seat is vacant, n the party identification of the last occupant is used. Given this framework, four different Senate situa- tions are derived: I) Party Challenge-~The RST and the incumbent are in the same party, and the incumbent is running for re~election (N=18). II) Party Succession—~The RST and the incumbent are in the same party, but the incumbent is not running for re~election (N=15). III) Opposition Party-~The RST and the incumbent are in different parties, and the incumbent is running for re~electioni(N=42). IV) Open-ended-~The RST and the incumbent are in different parties, but the incumbent is not running for re~election (N=10). Party Challenge (I) and Party Succession (II) are in-party situations (N=33), while Opposition Party (III) and Open-ended (IV) are out-party situations (N=52). Party Challenge (I) and Opposition Party (III) are situations in which the RST had to be concerned about facing the incumbent (N=60), while Party Succession (II) and Open-ended (IV) are situations in which the incumbent was not a candidate at the time of the RST's decision to try for the Senate (N=25). 124 See Table 27. TABLE 27 RSTs Classified by Senate-Incumbent's Candidacy and Party INCUMBENT INCUMBENT RUNNING NOT RUNNING TOTAL RST: Same Party I (N=18) II (N=l5) 33 Other Party III (N=42) IV (N=10) 52 Total 60 25 85 The single most interesting group to analyze are those who plan to challenge the incumbent of their own party (Group I). These Representatives may be viewed as the most ambitious. Certainly they are prepared to risk the most-~they risk their seat in the House (usually a safe seat) to run in a race where the odds of winning are slight. Only about 1 out of 10 Republican incumbents seeking renom~ ination are defeated, while the odds are 2 out of 100 for Northern Democrats, and a bit higher for Southern Democrats with 3 out of 10.99 These Representatives share the asset (or liability) a Senator has, of having a legislative record and not just a platform of promises to run on. But most do not share the Senator's statewide base of support. In fact only two of the eighteen (11%) in Group I had statewide elective experience, a lower proportion than the RSTs in any of the other three categories of Senate situations (these ranged from 17% to 99v.0. Key, Parties, Politics._and Pressure Groups, (5th edition; New York: Thomas I. Crowell Company, 1964), pp. “'42-’43. 125 50% with statewide experience). The ambition of these RSTs would lead us to hypothe- size several other characteristics. We would expect them to be younger and to have started a political career earlier and therefore not to be without political experience. They are not brash upstart Representatives out to build up a rep- utation for a subsequent race, nor are they oldtimers making a last-ditch fight. They probably have had political exper- ience prior to coming to the House, perhaps in the state legislature, suggesting a legislative experience ladder for them to climb. Their personal ambition suggests that it need not stem from the opportunity structure in the state. Their ambition also suggests that their interest in the Senate does not stem from a frustrated House career-~i.e. they would be expected to be from the better committees (even holding seniority constant). Their ambition, and early start suggests they are more likely than other RSTs to be lawyers and less likely to come from business occupations or agricultural occupations. These Representatives have had ample time to plan a campaign strategy, and changes in their behavior in the ultimate term would not be as likely among this group-- although a lack of statewide experience or organization may still result in behavioral changes prior to candidacy. Their ambition, longer planning period, and relatively safe House seats (in a relatively safe state)~~suggests that their specialization in the House might be low. But we would expect relatively high level of party cohesion, or at least 126 similar to the state party pattern. Although these hypotheses are stated to apply to all RSTs in this category, there are nevertheless several distinct groupings, and some differences may be expected. One sub-group is those who are risking their House seat to challenge an elected incumbent (there are nine in this group). One Representative (Dent) challenged an elected incumbent Senator, but simultaneously ran for re~election to the House; he obviously faced little risk, and his motives were other than personal ambition. Three Representatives (two in the same contest) challenged an incumbent who had been appointed to his position to fill a vacancy caused by the death of the elected Senator. These Representatives may have a shorter planning period than others in this category; and they clearly face less risk, as the appointed incumbent may not have built a statewide base of support either.100 Still another sub-group are the five Representatives who prepared to run against the incumbent, only to find the incumbent Senator withdrawing from the race. Although there may be little theoretical ground upon which to differentiate the latter group from the first group, there are neverthe- less some empirical differences. For example, those who gave up their House seat to run against the elected incum— bent Senator were younger and less experienced than those RSTs in Group I who did not actually face an incumbent in lOOV.O. Key exCIudes these from his calculations of what happens to incumbent Senators in their re~nomina~ tion bids; see ibid.. p. 438. 127 the primary. Representatives who run against incumbents of their own party may be expected to resemble RSTs in Group II on some characteristics, as they both are the "in-party". 0n the other hand, Group I and Group III RSTs share the impor- tant characteristic of planning to run against the incumbent, the former in a primary and the later in the general election. The data and analysis that follow will examine the relationship of the Senate situation and the selection of RSTs. Theoretically we assume that the Senate situation is the independent variable, and that if we find a preponder- ance of RSTs in a Group to have certain characteristics, we assume that the Senate situation tended to select or induce such Representatives to become Senatorial candidates. We will examine, therefore, the effect of the Senate situation upon agent characteristics (the personal and ambition attributes of the RSTs), origin characteristics (aspects of their House situation and behavior), and other destination characteris- tics (aspects of the situation surrounding the Senate seat they seek). The impact of the Senate situation upon the Repre- sentative's decision to try for the Senate may be seen by examining the proportion of contests in each Senate situa- tion which have RSTs competing. RST participation rates in contests varied directly with the degree of opportunity. They were least likely to be competing where the opportunity was the least-~i.e. where a candidate would have to make his opportunity by challenging the incumbent in the primary, or 128 Party Challenge (I) situations-~and most likely to be running when the opportunity was most favorable-~i.e. being a member of the in-party trying to succeed the incumbent who was not seeking re-nomination, or Party Succession (II) sit- uations. In the 1956-1966 period,101 only 1.8% of all pri~ maries in Party Challenge (1) situations had RSTs as candi- dates, compared to 36.8% of all Party Succession (II) primaries. The pattern was continued for the other two situation types. Assuming that the absence of an incumbent constitutes a greater opportunity for the out-party than if the incumbent were running, we would expect more Open-ended (IV) contests with RSTs than Opposition Party (III) contests; the data confirms this, with 27.8% of the former and 11.0% of the latter having RSTs running. Exclusion of uncontested primaries (or ones in which the winner had more than a 9~l lead over his nearest opponent) does not alter the conclusion; RST participation rates in these situations ranked by the level of opportunity were as follows: II:42.9%. IV:25.0%, III:10.1%. 1:3.92. Senate Situation and Agent Characteristics Representatives who plan to challenge the Senate incumbent of their own party were assumed to be the most 101This period was chosen because of the availability of data on Senate primaries in America Votes. Senate situa- tions for the purposes of this analysis were determined by the incumbent's candidacy in the primary; contests involving RSTs were reclassified to meet this definition, even if the RST had entered the contest before the incumbent withdrew. Nominations by conventions and other means were excluded. 129 ambitious group of RSTs. This assumption led us to expect younger men who started their political careers earlier than other RSTs. Tables 28 and 29 tend to confirm that they started their political careers earlier. But on several of the other measures linking age and political careers, it is hard to distinguish them from those in Group II~~Representa~ tives who share the Senator's party but are running after the latter's withdrawal from the race (by death or announce- ment of retirement). The age characteristics of the in-party RSTs (Groups I and II) tend to differ from those of the out- party RSTs (Groups III and IV). Their start in politics is earlier, and they have served longer before attempting to reach the Senate; they also reached the House earlier. But their attempt to become a Senator is not earlier than out- party RSTs. The differences within a group of RSTs are greater than differences between RSTs in different Senate situations, however. For instance, age at entry into public office ranged from 23 to 45 in Group I RSTs, 23 to 47 for Group II RSTs, 22 to 60 for Group III, and 23 to 51 for the ten RSTs in Group IV. The range of years in public office was from 5 to 33 in Group I, 9 to 30 in Group II, 2 to 27 in Group III, and l to 29 in Group IV. Similarly, we find a wide range of offices serving as their initial public office. RSTs in each of the four groups began their political careers in each of the types of offices considered, with the exception of statewide elective. But there were some differences in the 130 TABLE 28 Senate Situation and Age at Political Career Points I II III IV Entered First Public Office Before 30 10 56% 7* 50% 13 31% 4 40% Entered House of Reps. Before 40 8 44 9 60 12 29 4 4O Tried for Senate Before 50 8 44 8 53 28 67 4 40 N=18 N=15 N=42 x=10 *Excludes Reifel for whom information on age at first office is unavailable. TABLE 29 Senate Situation and Mean and Median Age At Political Career Points .71 2.11.1.1. I! Mean Age at First Public Office 31 33* 34 32 Median Age at First Public Office 28 30* 33 31 Mean Age at House Entry 41 39 42 44 Median Age at House Entry 42 39 42 46 Mean Age at Senate Try 49 50 48 50 Median Age at Senate Try 50 48 47 51 Mean Years of Public Office Exp. 15 14* 11 14 Median Years of Pub- lic Office Exp. 15 16* 9 ll *Excludes Reifel for whom complete information on first office is unavailable. distribution of these offices among the four groups. Group I RSTs were slightly more likely than any other group to have begun their public office holding career in the state legislature, and they were less likely than the other groups 131 to have started in public offices related to the law enforcement field. But RSTs in Group IV had almost as high a proportion of state legislative beginners (40% compared to 44% of Group I)~~so the legislative ladder hypothesized above for Group I RSTs does not appear to be exclusive (nor strong). See Table 30. TABLE 30 Senate Situation and First Public Office EIRST PUBLIC OFFICE I II III IV State Legislature 8 44% 5 33% 9 21% 4 40% Law Enforcement 2 ll 3 20 12 29 3 30 Congress 4 22 4 27 12 29 l 10 Administration 3 l7 2 13 7 17 l 10 Local Government _1 6 _l 2 _§ 5 _l 10 Total 18 100% 15 100% 42 101% 10 100% There do not seem to be any strong patterns along in-party/out-party lines, nor the other matrix dimension (incumbent running or not running for re~election). There are such patterns, however, apparently linking occupation and Senate situation~~i.e. certain Senate situations seem to attract RSTs with certain occupational backgrounds. For instance, lawyers were less likely (and businessmen more likely) to be challenging incumbents re- gardless of their party than not to be. See Table 31. This avoidance of incumbents may reflect a percep- tion of an unwritten rule among professional politicians to avoid conflict with incumbents if possible. This form of Job protection is seen in the phenomenon of "incumbent reap- Portionment" when a legislature will draft new district 132 lines after a census with the objective of protecting incum- bents rather than benefitting their party. Another line of speculation, however, is that lawyers, as more "professional politicians", are more concerned about their careers and try to avoid the more competitive situations. Some evidence for this speculation may be found in the data on RSTs which sug- gest that lawyers were slightly less likely than non-lawyers to run for "competitive" or "hopeless" Senate seats, and slightly more likely to run in "safe" seats; lawyers also were far less likely than non-lawyers to be involved in highly competitive Senate nomination struggles. TABLE 31 Senate Situation and Occupation I II III IV OCCUPATION N=18 N=15 N=42 N=10 Law 8 44% ll 73% 19 "452’ 8 80% Business 5 28% 2 13% 15 36%. 2 20% Teaching 5 2 7 - Agriculture 3 l 6 2 Journalism 3 - 3 - Other Professional 1 - l - In view of the disproportionately large number of lawyers in the situations where Senate incumbents are not candidates, it is surprising to find that the proportion of RSTs with law enforcement public office experience is uni- form in three of the four groups (only in Group I does much less than one-third of the RSTs have such experience). See Table 32. It is interesting to note that those challenging incumbents are less likely to have had statewide elective 133 experience-~whether gained from At-Large elections to the House of Representatives or from election to state offices ~~than those running where the incumbent is not a candidate (15% of the former comparajto 32% of the latter). TABLE 32 Senate Situation and Prior Public Office Experience I II III IV PUBLIC OFFICE EXP. N=18 N=15 N=42 N=10 State Legislature 10 56%» 6* ‘40%' 13 31%' 6 60% Law Enforcement 4 22 5 33 13 31 3 30 Statewide Elective 1 '6 ~ - 1 2 l 10 AdminiStration 5 28 2 l3 8 l9 1 10 Local Government 1 6 l 7 4 10 2 20 Statewide Elective (incl. At-Large House of Reps.) 2 ll 3 20 7 l7 5 50 Senate Situation and Origin Characteristics When we examined the total distribution of Repub- lican RSTs as compared to that of a control group of Repub- lican Representatives, we concluded that those who came from larger cities (100,000 plus~~or even those over 40,000) were under-represented in the RST population, while those from rural areas (towns of less than 2500 population) were over- represented. When this distribution of residence is examined for the four Senate situations, we can see that Republican RSTs who challenge incumbents of their own party do not come from larger cities (the largest city was Rep. Ellsworth's Lawrence, Kansas with 33,000). Although none of the Republican RSTs who challenged incumbents of their own party came from cities larger than 40,000 or those which were even one-half the size of their 134 state's largest city, 24% to 60% of the Republican RSTs in the other Senate situations did (e.g. 38% of the Republicans in Group II situations came from cities over 100,000). There is a slight tendency for large city Republicans to avoid in- cumbents regardless of their party status (31% of all Repub- lican RSTs are in incumbent not running situations-~Group II and IV, but 50% of all large city Republican RSTs are in such races). Democratic RSTs from large cities, on the other hand, were over-represented while those from rural areas were under- represented. In contrast to the Republican RSTs, Democrats from large cities are more likely to seek out incumbents than to avoid them. Forty percent (40%) of the Democrats in Group I come from cities over 100,000, and 40% are from the largest city in their state; this contrasts to 29% of Group II RST Democrats. Incumbent opposing Democrats (Groups I and III) constitute 72% of Democratic RSTs, but 88% of large city Democratic RSTs are in that category. See Table 33. TABLE 33 Senate Situation and Size of City of Residence, by Party DEMOCRATS REPUBLICANS CITY SIZE I II III Iv I II III _Iv POPULATION N=10 p31“ N=21 555 1:8 wig _N=21 h: under 2,500 20% 0 0 20% 38% 25% 19% 0 2,500-40,000 40% 71% 38% 60% 63% 38% 52% 40% 40,000-100,000 0 0 19% 20% 0 0 10% 40% 100,000 & over 40% 29% 43% O 0 38% 19% 20% CITY SIZE AS PRO- PORTION OF STATE'S LARGEST CITY less than 10%* 50% 71% 24% 60% 63% 63% 71% 20% 10-49% 10% 0 29% 20% 38% 13% 5% 20% 50-99% 0 0 10% 0 0 0 10% 20% city is largest 40% 29% 38% 20% 0 25% 14% 40% 135 v.0. Key observed that the primary system over- represented members from a party's dominant area-~i.e. Democratic areas contribute a disproportionate number of Democratic nominees for statewide office. Apparently the presence of an incumbent, whether he is of the same party or not, has a similar effect. Races against incumbents brings out Democratic candidates from large cities, but discourage Republicans from that area. The relationship of the relative size of the Congres- sional District to the nature of the Senate situation is not as clear. Representatives from At-Large districts are dis- proportionately more likely to be in Group IV--out-party with incumbent not running--than in the other categories. See Table 34. TABLE 34 Senate Situation and Proportionate Size of Congressional District CD SIZE I II III IV 1-19% of state 8 44% 7 47% 23 55% 4 40% 20-79% of state 8 44 6 40 13 31 2 20 100% of state _2 ll _2 13 _§ 14 _4 40 (At Large) Total 18 99% 15 100% 42 100% 10 100% We had speculated that the Representatives who challenge incumbents of their own party are not likely to be newcomers to Congress. The data supports this hypothesis-- only 17% were in their first or second term, compared to 33% of the other in-party RSTs and 46% of the out-party RSTs. In-party RSTs attempting to succeed their own party's 136 Senator, may have had to wait patiently for the opportunity --and in fact more than half of them had six terms or more in the House. See Table 35. TABLE 35 Senate Situation and House Seniority SENIORITI I II III IV 1-2 terms 3 17% 5 33% 20 48% 4 40% 3-5 terms 9 50 2 13 13 31 5 50 6+ terms _6 32 _§ 53_ ‘_2 21 _l 10 Total 18 100% 15 99% 42 100% 10 100% Our Speculation above suggested that RSTs who chal- lenge the incumbent of their own party can be assumed to be very ambitious. This ambition, we suggested, did not stem from a frustrated House career. In order to investigate this hypothesis, we will examine the committee assignments of RSTs in each of the four Senate situations. See Table 36. From this table we can see that RSTs in Group I have the best committee assignments. Two-thirds of the RSTs in that Group had committee assignments in the top committees (Rules, Ways and Means, Appropriations, Foreign Affairs, and Armed Services). This compares to 60% for fellow in-party RSTs (Group II), 40% for Group III, and 30% for Group IV. More importantly, RSTs who oppose incumbents of their own party, do so from positions on the best committees--regard- less of seniority. For instance 4 of 6 (67%) Group I high seniority RSTs have top committee assignments: this compares to 63% of the high seniority Group II RSTs, 67% of Group 1113, and none of the Group IVs. Even Group 1's low 137 ”manmu mlm ma xuanoaomm .mEhmu oHoE no 0 ma kufluowcmm swan . w m finance ”meuou NuH ma xuauoWCCm 30A .zaoo oomacmflmmm ummnmws one ma nomad am 4% x66 66m . b I. \ONH EEOC c H N u a n u AHHHV mmmoufl 6 HH H H a m H H H mcerme ummaoH NOH o\mm o o 6 mos Sufi H u H - 6H N - AHHV mmmubHEEOU N 6 6 6 6 H N m H wcchma esH662 Mom Nos Now New 6 - N H NH 6 6 m a 6 H 6 NH 6 6 N AHV mmmuuHEEQU mcaxnmm ummswam 6:38. m m m 388. ..H. m m 338. ...H. m w. 338. "H. m m nemacmema. 11: mmbuHBEOU zuHMOHcmm %DHHOHcmm >uHHOflcmm Muauowcwm >H HHH HH H thuHHOMGCm ma .ucmacwwmm< mmuuHEEoo pom coflumduwm mumcmm om mqmM_ 383 He apparently was going to be the Kennedy forces' candidate for the Senate. He had been closeted with Kennedy aides and with state chairman Burns. who was a Kennedy man. Sutton reportedly had informed the Coalition for a Democratic Alternative that he could not support McCarthy because he was being presented as a one issue man. On March 19th he officially announced his candidacy for the Senate. But this candidacy was to receive several serious . k .\I. —v. blows. Not unexpectedly some upstate party leaders com- plained that a Negro Senatorial candidate would hurt them ( in local races: in addition Harlem did not warmly receive his candidacy either (his announcement had noted defensively that his candidacy for the Senate would not deter his dedi- cation to the black cause). The most serious blow. however. was the action of the Liberal Party leaders. On March 28th--two days before the March 30th Democratic state committee endorsing conven- tion--the Liberal Party's policy committee voted to endorse Javits for re~election to the Senate. If this recommendation were adhered to by the full state committee. it would be the first time in the party's 24-year history that it had endorsed a Republican for statewide non-judicial office. Javits received 18 of the 25 votes in the committee. Sutton received 4. and 3 voted to run their own candidate in November. Sutton immediately declared his disappointment and his decision to re-assess his candidacy. Several phone calls convinced Sutton that the Liberal Party's endorsement action had served to seriously dry up potential campaign 384 contributions for him. Despite predictions of victory from such supporters as Nassau County party chairman John English. Sutton announced on the morning of March 29th that he was quitting the race. The first assessments were that this withdrawal would help Rep. Pike (he was already supported by some Kennedy men such as former state chairman William McKeon). In fact Resnick issued a statement attacking the Kennedy machine for allying itself with "the most conserva- tive Democrat north of the Mason-Dixon Line--namely Otis Pike". But that same afternoon efforts were launched to drum up support for another Kennedy team member. Eugene Nickerson the Nassau County Executive. The next day. after a night of calls and conferences, the Democratic state committee endorsed Nickerson. giving him 58% of the vote. Rep. Resnick received enough support (largely from upstate areas) to get 30% of the vote. suffi- cient to run in the primary. Rep. Pike received the remain- ing 12% of the vote. Paul O'Dwyer had decided not to compete for the state committee endorsement. because he felt his support there was unrepresentative of his true strength which lay with the McCarthy Clubs organized throughout the city and state. The Liberal Party made its endorsement of Javits official the day after the Democratic committee met. Pike soon decided to drop out of the Senate race and return to his bid for re-election to the House. O'Dwyer collected the required number of signatures. and entered the primary. The three-man Senatorial primary race continued to 385 be effected by developments in the Presidential nomination struggle. President Johnson's withdrawal from the race the very next day obviously undercut Resnick's candidacy. Res- nick endorsed Humphrey. but the latter never entered the New York primary for convention delegates, and did not cam- paign in the state. Nickerson tried to separate his candidacy from Kennedy's by appointing a committee which included supporters of all of the presidential candidates. But Nickerson was Kennedy's candidate. and the assassination of the latter in California only two weeks before the New York primary seriously damaged his campaign. The primary was extremely close. Paul O'Dwyer received 275,900 (or 36.3%). only 18,200 more than Nickerson's 257,000 (33.9%). Resnick trailed not far behind with 229,900 or 29.8%. In the November three-man Senatorial election (the Conservative Party ran its own candidate). Javits' plurality was over one million votes. Analytical Summary: Resnick (D) and Pike (D) Senate situation: Republican incumbent ran for re- election (III). Senate competition: Democrats won 2 of previous 5 elections for President. Senator. and Governor (Competitive). Career pattern: Law Enforcement (30-39%). Nomination struggle competition: Convention: 58% to RST's 30% and RST's 12%: primary: last RST did not run. other RST ran third getting 30% to winner's 36% (C) (D). Nomination struggle source: Former city councilman won the primary; county executive led in the conven- tion and came in second in primary. 386 1968 Pennsylvania Dent Democrat T Rep. John Dent almost carried off what many thought to be impossible-~the defeat of the incumbent in the Demo- cratic primary. Sen. Joseph Clark had angered many party leaders in his twelve years in the Senate. Nevertheless Clark still won the endorsement of the Democratic State Policy Committee 43 to 0 with 5 abstentions. But as Clark's in opposition to President Johnson's Vietnam policy mounted. Rep. John Dent decided to enter the race. Dent (60) was considered to be one of the state's western party leaders; l he had served for 23 years in the state legislature and 10 years in the House of Representatives. Pennsylvania law permitted him to seek both the Senate nomination and the nomination for re-election to the House. This dual entry led some to believe his Senate entry was not a serious one. But Dent's reply to a request for him to leave the Senate race for the sake of party unity. was that he "was in the race to stay. that I would not make a travesty of political aspirations. that this was no time for levity." President Johnson's announcement not to seek re- election apparantly cooled off the bitter 1% month primary campaign. and Dent practically stopped campaigning for the Senatorial nomination. Several days before the primary. observers were quoted as not expecting Dent to get over 30% of the vote. But instead Dent polled about 47% and trailed Clark by about 50.000 votes. The split in the Democratic ranks did not heal. 387 and Clark was defeated by Rep. Schweiker. Analytical Summary: Dent (D) Senate situation: Democratic incumbent ran for re- election (I). Senate competition: Democrats won 2 of previous 5 elections for President. Senator. and Governor (Competitive). Career pattern: Law Enforcement (30-39%). Nominfitéon struggle competition: High-~53% to RST's 7 C . Nomination struggle source: Incumbent. 1968 Pennsylvania Schweiker Republican RSW Richard Schweiker faced no opposition in the Republican primary for the Senatorial nomination. He had also won the unanimous endorsement of the Republican State Committee for that position. Earlier speculation had men- tioned some opposition to Schweiker on the groundsthat both he and Pennsylvania's other Republican Senator both came from the Eastern part of the state (Scott was from Phila- delphia. while Schweiker's district included some of the Philadelphia suburbs). Such sentiment. according to the New York Times. centered around State Secretary of Internal Affairs. John Taber of Pittsburgh. The incumbent. Senator Joseph Clark. was not having as easy a time in the Democratic primary. The split in the Democratic ranks (among other factors) contributed to Schweiker's victory over Clark in the November election. Schweiker received 52% of the vote. drawing 282.000 more votes than Clark did. 388 Analytica;_8ummary: Schweiker (R) Senate situation: Democratic incumbent ran for re- election (III). Senate competition: Republicans won 3 of previous 5 elections for President. Senator. and Governor (Competitive). Career pattern: Administrative (30-39%). Nomination struggle competition: Unopposed (AA). BIBLIOGRAPHY Anderson. Lee F.. Watts. Jr.. Meredith W.. and Wilcox. Allen R. ’Legrslative Roll-Call Analysrg. Evanston: Northwestern University Press. 1966. Barber. James David. The Lawmakers: Recruitment and Adaptation to Legislative Life. New Haven: Yale University Press. 1965. Berelson. Bernard R.. Lazarsfeld. Paul F.. and McPhee. William N. Voting: A Study of Opinion Formation ;n a Presidenrral Campaign. Chicago: University of Chicago Press. 1954. Campbell. Angus. et.al. The American Voter. New York: John Wiley & Sons. Inc.. 1960. Carney. Francis. "The Rise of the Democratic Clubs in California." Cases on Party Organization. Ed. Paul Tiléet. New York: McGraw-Hill Book Company, 1963. 32- 3- Clapp. Charles L. The Congressman: His Work as He Sees 12. Garden City: Anchor Books. 1963. Clem. Alan L. The Nomination of Joe Bottum: Analysis of a Committee Decision to Nominate a United States Senator. Vermillion: Governmental Research Bureau. University of South Dakota. 1963. Clem. Alan L. "Popular Representation and Senate Vacancies." Midwest Journal of Political Science. X (February. 1966). 52-77. Ill Congressional Quarterly, 1949-1969. Crane. Jr.. Wilder. "A Caveat on Roll-Call Voting Studies of Party Voting." Midwest Journal of Political Science. IV (August. 1960). 237-49. Cresap. Dean R. Party Politics_rn the Golden State. Los Angeles: Haynes Foundation. 1954. David, Paul T., Moos. Malcolm. and Goldman, Ralph M. 389 390 Presidential Nominating Politics in 1252. Vol. IV: The Middle West. Baltimore: The Johns Hopkins Press, 1954. Davidson. Roger H. The Role of the Congressman. New York: Pegasus. 1969. Derge, David R. "The Lawyer as Decision-Maker in the American State Legislature." Journal of Politics. XXI (August. 1959). 426-31. Dexter, Lewis Anthony. "The Representative and His Dis- trict." New Perspectgyes on the House of Representa- ) tives. Eds. Robert L. Peabody and Nelson W. Polsby. Chicago: Rand McNally & Company. 1962. 3-29. Eulau. Heinz, and Sprague, John D. Lawyers in Politics: . A Study in Professional Convergence. Indianapolis: The Hobbs-Merrill Co.. 1964. r Fishel. Jeff. "Party. Ideology. and the Congressional Challenger." American Political Science Review, LXIII (December. 1969), 1213-32. Froman, Jr.. Lewis A. The Congressional Process: Strategies, Rules, and Procedures. Boston: Little, Brown and Com- pany, 1967. Froman, Jr., Lewis A. Congressmen and Their Constituencies. Chicago: Rand McNally & Company, 1963. Froman, Jr.. Lewis A. "A Realistic Approach to Campaign Strategies and Tactics." The Electoral Process. Eds. M. Kent Jennings and L. Harmon Zeigler. Englewood Cliffs: Prentice-Hall, Inc.. 1966, 1-20. Frost, Murray. "Roll Call Cohesion in the New Jersey State Legislature. 1956-1957." Unpublished M.A. Thesis, Michigan State University. 1959. Goldman, Ralph M. "The National Party Convention." Present Trends in American National Government. Ed. Alfred J. Junz. New York: Frederick A. Praeger, 1961. 31—47. Goodwin, George. "The Seniority System in Congress." Ameri- can Poritica;:Science Review. LIII (June. 1959), 412-36. Harris, Joseph P. California Politics. Stanford: Stanford University Press, 1961. Hedlund. Donald D.. and Friesma. H. Paul. "Legislative Perceptions of Public Opinion.'' Paper presented to the American Political Science Association Meeting, Los Angeles. 1970. Hess, Stephen. America's Political Dynasties from Adams 391 to Kennedy. Garden City: Doubleday & Co.. Inc.. 1966. Hill, A. Spencer. "The 1964 Election in New Mexico." West- ern Pglitrcal Quarterly, XVIII (June. 1965). 499-501. Hinckley, Barbara. "Incumbency and the Presidential Vote in Senate Elections: Defining Parameters of Subpresiden- tial Voting." American PolitrpalgScience Review, LXIV (September. 1970). 836-42. Irion. Frederick C. "The 1962 Election in New Mexico." West- ern Political Quarterly. XVI (June. 1963), 448-52. Jacobs, Herbert. "Initial Recruitment of Elected Officials in the U.S.--A Model." Journal of Politics, XXIV (November. 1962). 703-16. Jewell. Malcolm E.. and Patterson. Samuel C. The Legislative Prpcess in the United States. New York: Random House. 1966. Kessel, John H. "A Game Theory Analysis of Campaign Strate- gy." The Electoral Process. Eds. M. Kent Jennings and L. Harmon Zeigler. Englewood Cliffs: Prentice-Hall. Inc.. 1966, 290-304. Key. Jr.. V.O. American State Politics: An Introduction. New York: Alfred A. Knopf. 1956. Key. Jr.. V.O. Parties, Politics, and Pressure Groups. 5th ed. New York: Thomas Y. Crowell Company, 1964. Key, Jr.. V.O. The Responsrble Electorate: Rationality in Presidential Voting. 1936-1960. New York: Vintage BOOKS , 1966 0 Key. Jr.. V.O. Southern Politics in State and Nation. New York: Vintage Books, 1949. Kingdon, John W. Candidates for Office: Beliefs and Strategies. New York: Random House, 1968. Leuthold, David A. Electioneering_in a Democracy: Campaign —_‘ for Congress. New York: John Wiley & Sons. Inc.. 1968. Lieberman, Joseph I. The Power Broker: A Biographykof John ngBailey Modern Political Boss. Boston: Houghton Miflin Co.. 1966. Los Angeles Times, November 10. 1970. Masters, Nicholas A. "Committee Assignments in the House of Representatives." American Political Science Review, LV (June, 1961). 345-57. \ 392 S Matthews. Donald R. U Senators and Their World. New York: Vintage Books. 1960. Mezey. Michael L. "Ambition Theory and the Office of Congressmen." Journal of Politics. XXXII (August, 1970)! 563-790 Miller. Warren E. "Presidential Coattails: A Study in Political Myth and Methodology." Public Opinion Quarterly. XIX (Winter. 1955-56), 353-68. Miller. Warren E.. and Stokes. Donald E. "Constituency Influence in Congress." American Political Science a Review. LVII (March. 1963). 45-56. Miller, Warren E..and Stokes. Donald E. "Party Government and the Saliency of Congress." Public Opinion Quar- terl . XXVI (Winter. 1962). 531- . New York Times, 1949-1968. Oleszek. Walter. "Age and Political Careers." Public Opinion Quarterly, XXXIII (Spring, 1969). 100-102. Parks, Norman. "Tennessee Politics Since Kefauver and Reese: A 'Generalist' View." Journal of Politics, XXVIII (Febo. 1966). 141-680 . Prewitt. Kenneth. The Recruitment of Political Leaders: A Study of Citizen-Politicians. Indianapolis: The Bobbs-Merrill Company. Inc.. 1970. Saloma. III, John S. Congress and the New Politics. Boston: Little. Brown and Company. 1969. Sarasohn. Stephen B.. and Sarasohn. Vera H. Political Party Patterns:;n Michigan. Detroit: Wayne State University Press. 1957. Schlesinger. Joseph A. Ambition and Politics: Political Careers_;n the Unitgd States. Chicago: Rand McNally & Company. 1966. Schlesinger. Joseph A. How They Became Governor. East Lansing: Governmental Research Bureau. Michigan State University, 1957. Schlesinger. Joseph A. "Lawyers and American Politics: A Clarified View." Mgdwest Journal of Political Salaries... I (May. 1957). 26-39. "‘"""‘"‘—" Schlesinger. Joseph A. "Political Careers and Party Leadership." Political Leadership_in Industrialized Societies: Studies in Comparative Analysis. Ed. 393 Lewis J. Edinger. New York: John Wiley & Sons, Inc.. 1967, 266-93. Schlesinger, Joseph A. "The Structure of Competition for Office in the American States." Behavioral Science, V (July. 1960), 197-210. Schlesinger, Joseph A. "A Two-Dimensional Scheme for Classifying States According to Degree of Inter-Party Competition." American Political Science Review, XLIX (December. 1955). 1120-28. Seligman. Lester G. "Opportunity, Risk, Selection and Decision-Making: A Model of Political Recruitments." Paper presented to the American Political Science Association Meeting, Los Angeles. 1970. Seligman, Lester G. "Political Parties and the Recruitment of Political Leaders." Political Leadership in Industrialized Societies: Studies in Comparative Analysis. Ed. Lewis J. Edinger. NeWTYork: John Wiley & Sons. Inc.. 1967. 294-315. Shannon. W. Wayne. Party. Constituency and Congressional Voting: A Study_of Legislative Behavior in the United States House of Representatives. BatongRouge: Louisiana State University Press, 1968. Smith, Roland E. "Political Recruitment and Political Risk in an Oregon Election." Paper presented to the American Political Science Association Meeting. Los Angeles, 1970. Sorauf. Frank J. Political Parties in the American System. Boston: Little, Brown and Company. 1964. Soule, John W. "Future Political Ambitions and the Behavior of Incumbent State Legislators." Midwest Journal of Political Science. XIII (August, 1969). 439-54. Stuart, Roger Winship. Meet the Senators. New York: Macfadden Books, 1963. Tuttle. Jr.. Daniel W. "The 1964 Election in Hawaii." Western Political Quarterly. XVIII (June, 1965), 1181-85 0 Vines, Kenneth H. "Federal District Judges and Race Relations Cases in the South." Journal of Politics, XXVI (maYO 1964). 337‘570 Wahlke John C. "Behavioral Analyses of Representative Eddies." Essays on the Behavioral Study of Politics. Ed. Austin Ranney. Urbana: University of Illinois Press. 1962. 173-90. m E..“. ‘m- :2-1 - 394 Wahlke, John C.. et.al. The Leggslative System: Explora- tions in Legislative Behavior. New York: John Wiley & Sons. Inc.. 1962. Wilson. James Q. The Amateur Democrat: Club Politics in Three Cities. Chicago: University of Chicago Press, 19 2. Witmer. T. Richard. "The Aging of the House." Political Science Quarterly. LXXIX (December, 1964), 526-41. Zeller. Belle (ed.). American State Legislatures. New York: Thomas Y. Crowell Company. 1954. 3:. HICH mmmIt“