Ask the Scholar
Document scope · 1 page
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory.
For page-specific OCR and visual context, open one of the page chats.
Source Description
This file contains:
From Khachigian to Magruder RE: Analysis of political behavior: Prospects for '72. 54 pgs. [Subject: Campaign] [Memo], 2/19/1971
Scholar Source Context
Document identity
localId
26146506
label
WHSF: Contested, 48-17
core
doc
dtoType
document
citationUrl
pageCount
1
Source metadata
id
26146506
sourceUrl
contentType
document
title
WHSF: Contested, 48-17
description
This file contains:
From Khachigian to Magruder RE: Analysis of political behavior: Prospects for '72. 54 pgs. [Subject: Campaign] [Memo], 2/19/1971
citationUrl
collections
Richard M. Nixon's Returned Materials Collection
Contested Materials Files
imageCount
1
hasImages
yes
source
import
hasTranscription
no
Source extras
naId
26146506
levelOfDescription
fileUnit
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
de850edf2faa5419
ocrText
Richard Nixon Presidential Library
Contested Materials Collection
Folder List
Box Number
Folder Number
Document Date
No Date
Subject
Document Type
Document Description
48
17
2/19/1971
Campaign
Memo
From Khachigian to Magruder RE: Analysis
of political behavior: Prospects for '72. 54
pgs.
Wednesday, June 03, 2015
Page 1 of 1
MEMORANDUM
THE WHITE HOUSE
WASHINGTON
DETERMINED TO BE AN
ADMINISTRATIVE MARKING
February 19, 1971
E.O. 12065, Section 6-102
By
NARS, Date
CONFIDENTIAL
MEMORANDUM FOR:
MR. MAGRUDER
FROM:
KEN KHACHIGIAN
Dec
SUBJECT:
ANALYSIS OF POLITICAL
BEHAVIOR: PROSPECTS FOR '72
The attached analysis does not lend itself to any easy answers.
I'll approach it on two fronts. First, assuming the theory is
correct, what should we do to implement this suggestion? Second,
is the theory correct?
If party alignment will control this election as well as major
elections to follow, then we should embark immediately on the
survey work which will tell us just how that party realignment
could take place; i.e., what can we do to encourage people to
change from Democrat or Independent to Republican registration?
If survey research and analysis can give us the way to achieve
party realignment, I suggest we go in that direction. Given a
good, sound analysis of how to go about the realignment, it would
not be difficult to gear our political operation in that direction.
But! I don't think party alignment will determine the election of
1972. Moreover, our first concern is not party realignment or
making the party strong, but to re-elect Richard Nixon. Winning
that election will come through the conventional appeal to current
Republicans; an appeal on the record to Independents; and an appeal
to the disillusioned Democrats who feel their party has deserted them.
- 2
It is only after elections that major changes in party alignment
take place. A case in point is FDR's victory in 1932. It was after
the election that so many Americans turned away from the
Republicans to the Democrats. And they turned away largely
because of what they perceived Roosevelt doing for them. Hence,
the huge realignment of Black voters who before had been Republican.
Thus, if any realignment is going to take place, it will take place
only after Richard Nixon is re-elected and not before. It will take
place only if there are factors which encourage that type of re-
alignment; that is to say, if Republicans can demonstrate that they
offer more to the Nation than do the Democrats. That will depend
on issue-presentation and issue-resolution -- two processes which
are just beginning.
So, what I am suggesting is that we keep our eye on the doughnut
and not on the hole. There is going to be no great party realignment
before the decade is over. It will be a slow process, barring
historical aberrations. What we can do is communicate to the public,
via the President, that his reforms and policies amount to the kind
of fundamental change on which realignments are built. If we can
communicate this, then we might be able to start the slow process
of party realignment.
But the process is slow, and if we gear up an election based on the
thesis of party alignment, we're going to lose sight of the real goal
of 1972; that is, winning an election.
DETERMINED TO BE AN
2/19/71
ADMINISTRATIVE MARKING
CONFIDENTIAL
E.O. 12065, Section 6-102
By
NARS, Date
(
MEMORANDUM FOR MR. MAGRUDER
FROM: KEN KHACHIGIAN
SUBJECT: MNALYSIS OF POLITICAL BEHAVIOR: PROSPECTS FOR '72
The attached
analysis does not lend itself to any
easy answers. I'll approach it on two fronts. First, assuming
the theory is correct,
what should we do to implement this
suggestion. ? Second, Is the theory correct?
alignment
That If party will control this election as well as
major
elections to follow, then we should embark immediately
on
the survey work which will tell us just how that party realignment
could take
place; i.e., what can we do to enfourage
people to change from Democrate or Independent to Rephulican
7
registration.
If survey research and analysis can give us the way to
achieve
party realignment, I suggest we go in that direction.
Given a good, sound analysis of how to go about the realignment,
it would not be difficult to
gear our political operation in
that direction.
But! I don't think party alignment will
determine
the eleection of 1972. Mofeover, our first concern is not
party realignment or making the party strong, but to re-elect
Richard Nixon. Winning that election will come through the
conventional appeal to current Republicans; an appeal on the
record to Independents; and an appeal to the disillusioned Democrats
who feel their party has deserted them.
that
It is only after elections
major changes in party alignment
page 2
take place. A case in point is FDR's victory in 1932. It was after
the
election that so many Americans turned away from the
Republicans to the Democrats. And they turned away largely because
of what they perceived Roosevelt doing for them. Hence, the
huge realignment of Black voters who before had been Republican.
Thus, if any realignment is going to take place, it
will
take place only after Richard Nixon is re-elected and not before.
It will take place only if there are factors which encourage that
type of realignment; that is to say, if Republicans can demonstrate
that they offer more to the Nation than do the Democrats. That
will depend on issue-presantation and issue-resolutinn -- two
processes which are just beginning.
So,
what I am suggesting is that we keep our eye on
the doughnut and not on the hole. There is going to be no
great party realignment
before the decade is
over. It will be a slow process, barring historical aberrations (sp?).
What we can do is communicate to the public, via the President, that
his reforms and policies amount to the kind of fundamental change
on which realignments are built. If we can communicate this, then
we might be able to start the slow process of party realignment.
But the process is slow, and if we gear up an election based
on the thesis of party alignment, we're going to lose sight of
the real goal of 1972; that is, winning an election.
i
ANALYSIS OF POLITICAL BEHAVIOR: PROSPECTS FOR 172
Kevin Phillips postulates an emerging Republican majority while
Scammon-Wattenberg insist that Democratic dominance is very much
the continuing reality. Someone is wrong
the question is who ? and
why ?
How Americans have voted is well established and the literature
analyzing past voting patterns is in general agreement. How Americans
will vote is conjecture and the literature is divided. Trends can be
established and theories developed, but in the final analysis the result is,
at most, an educated guess, a guess which is necessarily predicated upon
assumptions about the importance of identifiable shifts in the political
behavior of key voting blocs. These assumptions, bowever, tend to ignore
the dynamics of a continuing political process. It is difficult, if not
impossible, to identify from raw statistical data and charted voting patterns
how the political behavior of Americans will be influenced by future events - -
which is to say that trends can be accelerated or reversed depending upon
events only dimly perceived at the moment of analysis.
There are a number of variables that influence political behavior, and it
is striking that most analysts tend to concentrate on only those that are
statistically ascertainable: the demographic, political and historical
variables that constitute the "social antecedents" of political behavior.
11
These variables can be identified with some precision and enable the
analyst to determine how Americans have voted and are likely to vote,
"all things remaining equal. 11 However, there is a second category of
variables that must be considered if the fundamental political question --
Why do Americans vote as they do? is to be answered, and these are the
"attitudinal determinants" of political behavior: the attitudes of voters
toward issues, candidates, and parties. If you can identify these attitudes,
you can answer the crucial "Why?" question. Moreover, you can identify
those factors most likely to determine whether all things will remain
"equal. 11 At this point, you can proceed to consider ways to change
critical attitudes that will in turn change political behavior; a process of
applying "programmed political stimuli. 11 The name of the game, after
all, is to change voting patterns, not record them.
The dynamics of this systematic analysis of political behavior is
schematically set forth at Attachment A.
I. SOCIAL ANTECEDENTS
Voting blocs are identified by reference to demographic data. Thus,
it is possible to speak of the middle-class Irish Catholic vote in New York
City because empirically we can identify this bloc from the analysis of
raw demographic data setting forth the economic level, ethnic origin,
and religious tradition of voters in New York City. It is possible to
cross-reference various demographic variables in order to determine
the types of factors that influence voting behavior. By comparing voting
habits of middle-class and working class Irish Catholics in New York City,
we are fairly safe in attributing voting differences to economic background
since the other variables are constant. We can also introduce additional
variablés such as education, sex, age, marital status. In such a fashion,
it is possible to identify with some certainty the decisive factors that
influence various courses of political behavior.
In addition to these types of demographic data, it is also possible
to determine with some empirical certainty the established political
patterns of particular groups such as the extent to which they identify
with a party, the degree to which they are loyal to that party (their
identity is fixed as opposed to temporary), and the degree to which they
participate in the voting process (voter turnout). From this information
it is possible to determine, for example, that Irish Catholics in New York
City in 1900 identified with the Democratic Party, had a strong sense of
party loyalty (i.e., their allegiance did not shift perceptively from
election to election), and had a high degree of process participation
(i.e., they voted heavily). On the other hand, from similar types of
information it is possible to determine that black voters in New York
City in 1966 identified with the Democratic Party, were loyal to party,
but had a low level of process participation.
To the demographic and political variables must be added a third
set if a complete picture of the social antecedents of political behavior
is to be established. These may be characterized as historical,
economic, and societal. Normative as opposed to empirical, this data
is more difficult to collect and analyze but it is no less important. Class
consciousness, cultural tradition, peer group norms, or historical
experience can be important factors in determining political behavior.
It is difficult, for example, to account for the political behavior of the
Eastern European Jewish community in New York City during the 1920s
without reference to such factors. These normative variables cannot be
determined from statistical tables published by the Census Bureau but
must be determined by reference to less precise measuring devices.
Analysis of the social antecedents of political behavior is the necessary
first step for a sophisticated understanding of American voting patterns,
but for most students of the political process it is also the last step. The
3
vast majority of statistical analyses of political behavior are limited to
answering the question, "Now do Americans vote Of intellectual
interest, it is of little practical value to the working politician because
it postulates a situation in which it is possible to predict the outcome
but not influence it. It is of little comfort to a working politician to
know how an election will turn out (particularly if the indicators suggest
it is going to work out to his disadvantage). What a candidate or his
manager wants to know is not the How but the Why of American political
behavior. If you can isolate those variables that determine voting
patterns, you have a chance to alter those variables and thus influence
political behavior. From the study of the social antecedents of political
behavior it is possible to determine how Americans vote, and because
they vote with such partisan regularity during time-frames of approxi-
mately 32 to 36 years each, it is not only possible to determine past
performance but also future probability. To alter these voting patterns,
however, it is necessary to learn why identifiable political groupings vote
according to a particular pattern, and to learn why, you must identify
the attitudinal determinants of political behavior.
II. ATTITUDINAL DETERMINANTS
Scammon & Wattenberg's "Dayton housewife" would be hard-pressed
to explain rationally the motives, interests, and values that influence her
4
political behavior, yet only by reference to these attitudinal determinants
is it possible to understand why she votes for Candidate X as opposed to
Candidate Y.
There are three crucial elements in an election campaign: (1) issues,
(2) candidates, and (3) parties. Voter attitudes toward these elements
are largely determinative of election results and it is necessary to identify
these attitudes if the question of why Americans vote as they do is to be
answered
A. ISSUES
The voter's attitude toward issues is largely the result of four variables:
perception, intensity, volatility, and political characterization. A question
of public policy is not a political issue unless the volers perceive it as one.
The desirability of floridation does not become a political issue mercly
because a candidate chooses to discuss it; if the voters do not perceive it
as an issue, do not regard it as a serious question of personal or public
concern, it is a non-issue and of little influence in affecting voting patterns.
On the other hand, the failure of a candidate to discuss a question does not
rule out that question as an issue in the campaign. If the voter perceives
the question of unemployment as an issue, the candidate who fails to
address himself to that question does not climinate the issue, he only
avoids it.
5
The relative importance of an issue in a campaign is determined by
the intensity with which the voter identifies it as an issue. Rural Yankees
may regard abortion as a legitimate political issue, but not attach a great
deal of importance to it. Urban Catholics, however, may attach a great
deal of importance to it; for them it is a major issue. The difference
in attitudes toward such an issue is analytically determinable by gauging
the intensity of perception.
Some issues are highly controversial or volatile. The race question,
for example, may be a slumbering issue of only minor importance until
a ghetto riot breaks out and suddenly the issue is perceived with a high
degree of intensity as a major factor in the campaign. The volatility of
an issue is a red flag to a candidate to be on the alert for sudden develop-
ments that may turn a minor issue into a major one overnight.
Finally, and perhaps most importantly, the manner in which the voter
politically characterizes an issue is a major attitudinal determinant of
voting behavior. Unemployment, for example, is associated by many
voters with the Republican Party. For such voters it is natural to
support a Democrat in a campaign where unemployment is perceived
to be a major issue.
Issues assume importance in a campaign in proportion to their
influence on political behavior. The degree of this influence is largely
explainable in terms of voter attitudes toward the issues, and thus a
candidate who seeks to influence political behavior must first be aware
of what the voters perceive to be issues, how intense this perception
is, and the degree to which the voter has politically characterized the
issue as associated with a particular party.
B. CANDIDATES
The Dayton housewife's attitude toward a candidate is determined by
her affective disposition toward him, her objective appraisal of him,
and her association of him with the party or interests with which she
identifies.
Affective disposition toward a candidate is simply the emotional
response to a man. A good example, of course, is Jack Kennedy who
had great emotional appeal to significant voting blocs. Affective
disposition is a key attitudinal determinant of most behavioral patterns;
how one "feels" about another is often determinative of what one "thinks"
about him. Political attitudes are more often determined by emotional
factors than democratic theory postulates, and it is therefore necessary
for a political analyst to determine as precisely as possible the degree
to which voting is influenced by emotional reaction to the candidates
and what accounts for that reaction.
Objective appraisal of a candidate encompasses the rational,
dispassionate assessment of the man; a feat that occurs with singular
7
irregularity. However it does occur, and it is important. It was
particularly important for the President during the 1968 campaign when
many voters who were not affectively disposed toward him were able
(for reasons largely attributable to some of the variables outlined
above) to objectively appraise him and render a substantively rational
electoral judgment.
Of critical importance, particularly to a Republican candidate for
national office, is the degree to which the voter associates the candidate
as hostile to the party or interests with which the voter identifies. A
voter who has a strong sense of loyalty to the Democratic Party is
likely to be influenced in his attitude toward a Republican candidate
merely because of the candidate's party label. Likewise, a voter who
identifies his own interests with the welfare of farmers is likely to be
influenced in his voting by his association of a candidate with urban
interests. This associational determinant is important for it is not so
much a judgment of the candidate as a man (either emotional or rational)
as it is a judgment of the candidate as a representative. Attitudes toward
a candidate, therefore, are determined at two levels of perception: that
of man qua man and that of man qua symbol or representative.
8
C. PARTIES
The final significant electoral factor is party identification. Parties
assume particular significance in our electoral system because of our
historical experience with one party domination. Party loyalties, once
set, tend to remain set. For this reason, party identification is perhaps
the most important single determinant of voting behavior.
Beliefs and stereotypes about parties are a major attitudinal determinant
of political behavior. In the South, Democratic loyalties remain strong
because of the force of tradition and the stereotype of the GOP as the Party
of Reconstruction (this is changing, of course, in national elections, but
not too much in state and local elections). In working class neighborhoods,
antagonism against the GOP is rooted in the belief that the Republicans
are the Party of Big Business. These are powerful attitudinal determinants,
rooted as they are in deeply held belief and well established stereotype.
A less emotional, and hence less decisive, attitudinal determinant is
the degree to which the voter identifies party with issues and interests of
importance to him. The South is the best example of this determinant at
work, for although hostility to the GOP is traditionalized by stercotype,
a perceptible shift in party loyalty has occurred as a result of voters
identifying the Republican Party nationally with those issues and interests
9
of most importance to them. Party loyalties are firm, but not fixed,
and they shift as issue/interest identification begins to overcome beliefs
and stereotypes.
Finally, attitudes toward parties can be influenced by the suscepti-
bility of voting blocs to organizational efforts. This statement is merely
a verbal elaboration of the observable practice of systematic recruiting
of people into the party through organizational techniques. Slow, tedious,
and difficult, voter attitudes toward parties can be changed through
conscious organizational efforts. However, of all the determinants of
political behavior, organizational susceptibility is probably the least
significant, at least in short-range terms.
III. PROGRAMMED POLITICAL STIMULI
To the working politician, the study of social antecedents and
attitudinal determinants of political behavior is more than an intellectual
exercise, an accounting of how and why Americans vote as they do; it
is an indespensible prerequisite for a sophisticated campaign designed to
change voting patterns by influencing political behavior.
From data setting forth the social antecedents of political behavior
it is possible to determine who votes and how they vote. Such data is
useful in identifying those groups which are most likely to support a
candidate and, once identified, special attention can be devoted to them.
10
However, what a candidate really wants to know is how he can reach
voters not already disposed to support him, which means he wants to
know why people vote as they do and what can be done to change their
initial disposition.
Attitudinal determinants of political behavior are the key to this
vote-changing process: change attitudes and you change votes. Thus,
a well managed campaign is one in which a conscious effort is made to
alter attitudes by the application of programmed political stimuli.
If issues, candidates, and parties are the key elements in the
election process and attitudes toward these elements are central to
voting behavior, it is logical that the candidate's objective is to approach
each elément with an cye on its implications for voter attitudes. Issue
development, candidate image, and party emphasis become the keys to
electoral success.
A. ISSUES
The first step is to identify which issues are important to which voting
blocs and how important they are to each. There are bound to be conflicts
among voting blocs so that it becomes necessary to make a choice as to
which bloc shall be appealed to on the basis of a particular issue. In
making this choice, it may be that an issue important to Catholics can
only be emphasized by alienating Southern Baptists; however, it may be
such alienation can be avoided by also emphasizing another issue
important to Southern Baptists but of little interest to Catholics; that
is, decisions on issue emphasis require more imput than the attitude
of contending blocs toward the single issue in question.
Some voting blocs will perceive an issue to be important which a
candidate can not develop either because it will clearly cost him too
much with other important blocs or because the candidate is in an
inherently weak position on the issue. The best example of the latter
case was the economy during the 1970 campaign; we had to avoid the
issue because of the inherent weakness of our position. Knowing which
issues to emphasize and which issues to avoid is often as important as
knowing which issues should be developed in order to maximize strength
among a particular segment of the voting population.
Once a decision is made to develop an issue, the question is whether
it should be developed rhetorically or programmatically. This is a
major consideration for an incumbent who is in a position to deliver
programmatic solutions to issues.
Rhetoric is persuasive and profitable only when it addresses an issue
of concern to the voters in a manner that is convincing and reassuring.
Issue rhetoric must capture the aspirations of the electorate, must be
12
credible, and must hold promise of being translated into concrete
results. Rhetoric is not a substitute for programmatic development
of issues; it is a supplement, at most a holding action. Issue rhetoric
must be distinguished from image rhetoric -- the latter implicitly
involves nothing more than reassurance and confidence, while the
former necessarily implies action and resolution of issues in dispute.
Being able to determine when rhetoric is a sufficient response to the
issuc-oriented concern of the electorate is an art, not a science, and
is most difficult. IL is, however, also most important, particularly
when a party realignment is in progress.
If political behavior is to be altered by influencing the attitudes of
voters toward political issues, it is imperative that those issues be
defined in reference to particular voting blocs, that the attitudes be
identified, and that the response, rhetorical or programmatic, be of
such force as to have a major impact on those attitudes. Some issues
must and should be avoided, but not out of ignorance of the fact that
some voters regard them as issues. Issue avoidance should be a
calculated political decision based on an assessment that the candidate
is inherently disadvantaged on a particular issue. This disadvantage can
possibly be compensated for by development of another issue of equal
13
concern to the targeted voting bloc or by concentration on altering
that bloc's attitude toward candidate or party; the value of a systematic
analysis of the variables that influence political behavior is that it
identifies a variety of ways by which voter attitudes can be changed.
B. CANDIDATE IMAGE
No one need tell a candidate that his image before the electorate is
a decisive factor in election results. However, it does need to be
emphasized that in terms of party realignment leading to one party
domination of the political process, candidate image does not play as
important a role as issue development and party emphasis. Lincoln and
Roosevelt came and went, but the party they led remained in power long
after they had disappeared from the political scene. Candidate image
is most important in critical and deviant elections the former being
exemplified by the critical victories of Jefferson, Jackson, Lincoln and
Roosevelt, the latter by the deviant victories of Harrison, Taylor,
Cleveland, Wilson, and Eisenhower. Only history will tell if the Nixon
victory was critical or deviant.
Leadership (or candidacy) is generally characterized as charismatic
or institutional. The critical distinction between the two types may be
simply the manner in which the leader is perceived by the public: either
14
emotionally or objectively, as an unusual man whose influence is
independent of his institutional power or position, or as a more traditional
figure whose influence is dependent upon his position in or his identity
with the institutional order.
Although it is doubtful that charisma can be artificially created, there
is reason to believe that it can be artificially enhanced; a rough diamond
can be polished, but a leopard can't change its spots (to coin a phrase).
The point, of course, is that a candidate whose appeal is essentially
institutional cannot be repackaged and marketed as a charismatic leader
without creating a credibility problem.
There is no reason to believe that charismatic leadership is inherently
superior or more successful than institutional leadership. It is essentially
a difference of style, of technique, and it is by results that history judges
the success of leadership. A non-charismatic candidate who frets about
the absence of such appeal often makes a serious mistake by attempting
to project a pseudo-charismatic image at odds with a natural image
popularly more credible and electorally more helpful. Who would have
believed General Eisenhower in Camelot?
Image, presidential and otherwise, has been over-emphasized because
misunderstood. In a stable political order characterized by a two party
15
system, image, unless overtly offensive, is essentially a neutral
factor, less important in influencing political behavior than issues
and party. The candidate of a majority party can normally rely upon
the loyalty of his party members to carry him to victory; his only
concern is that his image not jeopardize the existing electoral fidelity
of this majority that identifies with his party. This was Lyndon Johnson's
problem. He represented the dominant political party, and in the course
of normal events, party loyalty should have been sufficient to guarantee
his re-election. However, he projected a distinctly negative image that
alienated members of his own party. Additionally (and perhaps more
importantly) his handling of the issues (race and war) alienated many
who had previously identified with his party. Lyndon Johnson didn't
need charisma to win in 1964; he would have had to have charisma plus
a bonus of good luck to win in 1968.
Candidate image is only one of three factors that influences political
behavior, and it should be considered in this limited scope. That is, it
ought to be considered in reference to the precise ways in which it can
alter political attitudes.
Affective disposition or emotion is the principal determinant of
attitudes toward a candidate and where this element is identified, steps
must be taken to deal with it. The "Tricky Dick" image, for example,
16
was such a problem. Having identified it as a problem, it was not
difficult to take positive steps designed to dispell the image.
An overlooked "image" problem relates to the manner in which key
voting blocs identify a candidate with interests they deem contrary to
their own. Thus, Lithuanians who believe a candidate is too soft on
the Soviets can be reassured by a statement to the contrary, an
appearance at an event of major interest to the Lithuanian community,
or by other gestures designed to correct the image. Likewise, union
members who believe a candidate is too closely associated with manage-
ment can be reassured by steps which are "image corrective" in nature.
Image is more than what comes across on a television screen. It is
principally a matter of perception by members of key voting blocs, and
a particular image that is favorable with one bloc may not be with
another. It is necessary therefore to identify existing image problems and
take steps to correct them. Generally they can be corrected without
generating a new image problem with another voting bloc.
Although mass communications have tended to nationalize many social
and political attitudes, they have not climinated all attitudinal distinctions.
As long as different political blocs with different views and attitudes exist
(that is, as long as America remains a pluralistic society), there will
be differences of image perception among the public, and no national
candidate can be satisfied with projecting a strictly "national" image.
C. PARTY EMPHASIS
In early August of 1968, it was widely believed that il was worthless
to consider past voting patterns as an indicator of future political behavior
because events were proceeding at such a rapid pace with such unexpected
results that voter attitudes were influenced only by the latest political
stimuli. At the same time, however, some skeptics were suggesting in
the face of polls showing Nixon with an overwhelming lead that by election
day traditional voting habits would begin to assert themselves, that party
loyalty would once again prove to be the most decisive (but not necessarily
the only decisive) determinant of voting behavior.
It is now generally recognized that this skeptical view was the most
practical and most accurate one. In the final weeks of the campaign
disillusioned Democrats who had toyed with the idea of voting for Wallace
or for Nixon returned to the ranks of their party. This was particularly
true of blue collar workers and others subject to the influence of organized
labor.
In spite of the rapidity of social and political change, in spite of the
obvious dissatisfaction with their party, its candidates, and its position
on the issues, large numbers of Democrats decided in the final days of the
campaign to stay with their party. A persuasive case can be made that the
18
remarkable thing about 1968 was not the number of Democrats who
bolted, but the number who stayed aboard what had all the appearances
of a sinking ship. One can and must conjecture about what would have
happened to the Wallace vote if the Governor had pulled out, but one
would be naive in the course of doing so to discount the likelihood that
traditional party loyalty would have been the decisive factor in its
distribution.
The point, of course, is that identification with party is and always
has been the principal determinant of political behavior. We identify
some elections as "deviant" because of their conspicuousness as
exceptions to the pattern of political control by a dominant party. Until
Woodrow Wilson's successful bid for re-clection, no minority party was
able to elect its candidate to the Presidency for two successive terms.
Cleveland might have accomplished this feat if his electoral vote in 1888
had reflected his popular vote and he was elected to a second term in 1892.
However, it is often overlooked by casual students of this period that the
Democrats were by the mid-80s rapidly becoming the majority party; were
it not for the Depression of 1893 and the seizure of the party by the
populists in 1896, the Democrats very likely would have become the
dominant party while Franklin D. Roosevelt was still a school boy. The
election of 1896 was a critical one for the GOP.
19
Without taking exception to Kevin Phillips analysis, candor requires
the observation that the most discouraging aspect of the prospects for
the 1972 election is the fact that available evidence suggests that the
GOP remains, by a substantial margin, the minority party. Our
posture much more clearly parallels that of Woodrow Wilson in 1916
than it does Franklin D. Roosevelt in 1936. For this reason, the
attitude of voters toward party will be critically important in 1972 and
a fundamental and crucial decision will have to be made about the
degree to which the Republican Party will be emphasized as the vehicle
through which the aspirations of the electorate can be realized.
This will be a difficult decision because a major party alignment is
4
not achieved by virtue of personalities. Ike didn't do it and the
President can't do it. If a realignment of the sort that Kevin postulates
is to take place, it must occur as the result of a conscious decision by
the voters that their interests are most clearly served by the Republican
as opposed to the Democratic Party and the President must run as a
standard bearer and not on his own. If, however, it is clear that we
are not in the process of such a realignment (or that we are not yet far
enough along in the process for it to be finally consummated in 1972),
the President's identification with the GOP may prove to be a disadvantage.
20
The dilemma we face is not simply one of gauging the degree of voter
identification with the Republican Party. If we assume that we are in a
1916-type situation in which party should be de-emphasized and adopt that
course, we may in the process retard what could be a significant shift in
party realignment. On the other hand, if we assume that we are in a
1936-type position in which we can move in for the kill and establish GOP
domination, we may jeopardize the President's re-election chances if we are
wrong.
What is required, therefore, is some precise study and analysis of
voter attitudes toward party. To be helpful, this study and analysis must
be centered on special voting blocs in special parts of the country. Gallup
polls which purport to show the party breakdown nation-wide are not very
helpful. What we are interested in is party attitudes among critical voting
blocs whose support is imperative for a party realignment to take place.
National shifts are interesting, but not particularly important. Critical
elections occur in key geographical areas among key voting blocs; if they
shift, it won't be long before others follow.
It is not enough, of course, to identify present attitudes toward party.
What is required is to identify what can be done to change these attitudes.
This process relates almost exclusively to the development and presentation
of issues. Ikc carried Michigan and Virginia; the former because of
21
personality, the latter because of issues, Only Virginia represents an
example of party realignment favorable to the GOP, and it should be
remembered that over the long-term only issue/interest identification
can permanently overcome party stereotypes which is to say that
candidate image may elect a President, but issues establish dominant
party control.
IV. FROM THEORY TO PRACTICE
The above discussion suggests a conceptual framework for the analysis
of political behavior for the purpose of altering voting patterns. It
presupposes that the variables that determine political behavior can be
identified and, once identified, influenced. It is not a mysterious process
but merely the systemization of what every practicing politician attempts
to do intuitively. It calls for the application of modern technology and
advanced political theory to the practical problems of a Presidential
campaign.
Demographic and voting statistics are readily available and, in machine
readible form, can easily be handled by computers. From such data
sophisticated analysts can readily identify significant attitudinal determinants
of political behavior that can be influenced by programmed political stimuli.
22
The significant advantage of this type of program is that it provides
more precise and reliable information upon which to base sound
political decisions; it clevates the "hunch" to a fact. There is no
substitute for good political judgment --- which means that the crucial
element in any process of political analysis is the analyst - but the
quality of analysis also depends upon the quality of the information
upon which it is based. Give a good analyst good information and you
are ahead of the game. We can bc, if we choose to be.
TYPES OF VARIABLES RELEVANT TO THE ANALYSIS OF POLITICAL BEHAVIOR
I.
discovient, Economic, Societal
Political
Metorical experionce/heritage
party identification
& religious tradition
SOCIAL
party regularity
family/peer group norms
ANTECEDENTS
process participation
class conscicusness
Parties
beliefs and
Democraphic
II
stereotypes
issue/interest
sox, ago, marital
identification
status
ATTITUDINAL
organizational
race/ethnic origin
meligion
Issues
DETERMINANTS
susceptibility
Motives, interests,
economic status
perception
education
intensity
values]
Candidates
place of residence
volatility
affective dis-
political charac-
terization
position
POLITICAL
objective appraisal
BEHAVIOR
party/interest
association
Printy Emohasis
(de-emphasis
PROGRAMMED
POLITICAL
STIMULI
Pattern of
forme Do Divelopment
Communication
Candidate Image
photorist /programmatic ]
charismatic/institutional
KLK
ANALYSIS OF POLITICAL BEHAVIOR: PROSPECTS FOR '72
Kevin Phillips postulates an emerging Republican majority while
Scammon-Wattenberg insist that Democratic dominance is very much
the continuing reality. Someone is wrong -- the question is who? and
why ?
How Americans have voted is well established and the literature
analyzing past voting patterns is in general agreement. How Americans
will vote is conjecture and the literature is divided. Trends can be
established and theories developed, but in the final analysis the result is,
at most, an educated guess, a guess which is necessarily predicated upon
assumptions about the importance of identifiable shifts in the political
behavior of key voting blocs. These assumptions, however, tend to ignore
the dynamics of a continuing political process. It is difficult, if not
impossible, to identify from raw statistical data and charted voting patterns
how the political behavior of Americans will be influenced by future events
which is to say that trends can be accelerated or reversed depending upon
events only dimly perceived at the moment of analysis.
There are a number of variables that influence political behavior, and it
is striking that most analysts tend to concentrate on only those that are
statistically ascertainable: the demographic, political and historical
variables that constitute the "social antecedents" of political behavior.
These variables can be identified with some precision and enable the
analyst to determine how Americans have voted and are likely to vote,
"all things remaining equal. 11 However, there is a second category of
variables that must be considered if the fundamental political question - -
Why do Americans vote as they do? -- is to be answered, and these are the
"attitudinal determinants" of political behavior: the attitudes of voters
toward issues, candidates, and parties. If you can identify these attitudes,
you can answer the crucial "Why?" question. Moreover, you can identify
those factors most likely to determine whether all things will remain
"equal. " At this point, you can proceed to consider ways to change
critical attitudes that will in turn change political behavior; a process of
applying "programmed political stimuli. 11 The name of the game, after
all, is to change voting patterns, not record them.
The dynamics of this systematic analysis of political behavior is
schematically set forth at Attachment A.
I. SOCIAL ANTECEDENTS
Voting blocs are identified by reference to demographic data. Thus,
it is possible to speak of the middle-class Irish Catholic vote in New York
City because empirically we can identify this bloc from the analysis of
raw demographic data setting forth the economic level, ethnic origin,
and religious tradition of voters in New York City. It is possible to
cross-reference various demographic variables in order to determine
the types of factors that influence voting behavior. By comparing voting
habits of middle-class and working class Irish Catholics in New York City,
we are fairly safe in attributing voting differences to economic background
since the other variables are constant. We can also introduce additional
variables such as education, sex, age, marital status. In such a fashion,
it is possible to identify with some certainty the decisive factors that
influence various courses of political behavior.
In addition to these types of demographic data, it is also possible
to determine with some empirical certainty the established political
patterns of particular groups such as the extent to which they identify
with a party, the degree to which they are loyal to that party (their
identity is fixed as opposed to temporary), and the degree to which they
participate in the voting process (voter turnout). From this information
it is possible to determine, for example, that Irish Catholics in New York
City in 1900 identified with the Democratic Party, had a strong sense of
party loyalty (i. , their allegiance did not shift perceptively from
election to election), and had a high degree of process participation
(i. e., they voted heavily). On the other hand, from similar types of
information it is possible to determine that black voters in New York
City in 1966 identified with the Democratic Party, were loyal to party,
but had a low level of process participation.
To the demographic and political variables must be added a third
set if a complete picture of the social antecedents of political behavior
is to be established. These may be characterized as historical,
economic, and societal. Normative as opposed to empirical, this data
is more difficult to collect and analyze but it is no less important. Class
consciousness, cultural tradition, peer group norms, or historical
experience can be important factors in determining political behavior.
It is difficult, for example, to account for the political behavior of the
Eastern European Jewish community in New York City during the 1920s
without reference to such factors. These normative variables cannot be
determined from statistical tables published by the Census Bureau but
must be determined by reference to less precise measuring devices.
Analysis of the social antecedents of political behavior is the necessary
first step for a sophisticated understanding of American voting patterns,
but for most students of the political process it is also the last step. The
3
vast majority of statistical analyses of political behavior are limited to
answering the question, "How do Americans vote?" Of intellectual
interest, it is of little practical value to the working politician because
it postulates a situation in which it is possible to predict the outcome
but not influence it. It is of little comfort to a working politician to
know how an election will turn out (particularly if the indicators suggest
it is going to work out to his disadvantage). What a candidate or his
manager wants to know is not the How but the Why of American political
behavior. If you can isolate those variables that determine voting
patterns, you have a chance to alter those variables and thus influence
political behavior. From the study of the social antecedents of political
behavior it is possible to determine how Americans vote, and because
they vote with such partisan regularity during time-frames of approxi-
mately 32 to 36 years each, it is not only possible to determine past
performance but also future probability. To alter these voting patterns,
however, it is necessary to learn why identifiable political groupings vote
according to a particular pattern, and to learn why, you must identify
the attitudinal determinants of political behavior.
II. ATTITUDINAL DETERMINANTS
Scammon & Wattenberg's "Dayton housewife" would be hard-pressed
to explain rationally the motives, interests, and values that influence her
4
political behavior, yet only by reference to these attitudinal determinants
is it possible to understand why she votes for Candidate X as opposed to
Candidate Y.
There are three crucial elements in an election campaign: (1) issues,
(2) candidates, and (3) parties. Voter attitudes toward these elements
are largely determinative of election results and it is necessary to identify
these attitudes if the question of why Americans vote as they do is to be
answered.
A. ISSUES
The voter's attitude toward issues is largely the result of four variables:
perception, intensity, volatility, and political characterization. A question
of public policy is not a political issue unless the voters perceive it as one.
The desirability of floridation does not become a political issue merely
because a candidate chooses to discuss it; if the voters do not perceive it
as an issue, do not regard it as a serious question of personal or public
concern, it is a non-issue and of little influence in affecting voting patterns.
On the other hand, the failure of a candidate to discuss a question does not
rule out that question as an issue in the campaign. If the voter perceives
the question of unemployment as an issue, the candidate who fails to
address himself to that question does not eliminate the issue, he only
avoids it.
5
The relative importance of an issue in a campaign is determined by
the intensity with which the voter identifies it as an issue. Rural Yankees
may regard abortion as a legitimate political issue, but not attach a great
deal of importance to it. Urban Catholics, however, may attach a great
deal of importance to it; for them it is a major issue. The difference
in attitudes toward such an issue is analytically determinable by gauging
the intensity of perception.
Some issues are highly controversial or volatile. The race question,
for example, may be a slumbering issue of only minor importance until
a ghelto riot breaks out and suddenly the issue is perceived with a high
degree of intensity as a major factor in the campaign. The volatility of
an issue is a red flag to a candidate to be on the alert for sudden develop-
ments that may turn a minor issue into a major one overnight.
Finally, and perhaps most importantly, the manner in which the voter
politically characterizes an issue is a major attitudinal determinant of
voting behavior. Unemployment, for example, is associated by many
voters with the Republican Party. For such voters it is natural to
support a Democrat in a campaign where unemployment is perceived
to be a major issue.
Issues assume importance in a campaign in proportion to their
influence on political behavior. The degree of this influence is largely
explainable in terms of voter attitudes toward the issues, and thus a
candidate who secks to influence political behavior must first be aware
of what the voters perceive to be issues, how intense this perception
is, and the degree to which the voter has politically characterized the
issue as associated with a particular party.
B. CANDIDATES
The Dayton housewife's attitude toward a candidate is determined by
her affective disposition toward him, her objective appraisal of him,
and her association of him with the party or interests with which she
identifies.
Affective disposition toward a candidate is simply the emotional
response to a man. A good example, of course, is Jack Kennedy who
had great emotional appeal to significant voting blocs. Affective
disposition is a key attitudinal determinant of most behavioral patterns;
how one "feels" about another is often determinative of what one "thinks"
about him. Political attitudes are more often determined by emotional
factors than democratic theory postulates, and it is therefore necessary
for a political analyst to determine as precisely as possible the degree
to which voting is influenced by emotional reaction to the candidates
and what accounts for that reaction.
Objective appraisal of a candidate encompasses the rational,
dispassionate assessment of the man; a feat that occurs with singular
irregularity. However it docs occur, and it is important. It was
particularly important for the President during the 1968 campaign when
many voters who were not affectively disposed toward him were able
(for reasons largely attributable to some of the variables outlined
above) to objectively appraise him and render a substantively rational
electoral judgment.
Of critical importance, particularly to a Republican candidate for
national office, is the degree to which the voter associates the candidate
as hostile to the party or interests with which the voter identifies. A
voter who has a strong sense of loyalty to the Democratic Party is
likely to be influenced in his attitude toward a Republican candidate
merely because of the candidate's party label. Likewise, a voter who
identifies his own interests with the welfare of farmers is likely to be
influenced in his voting by his association of a candidate with urban
interests. This associational determinant is important for it is not so
much a judgment of the candidate as a man (either emotional or rational)
as it is a judgment of the candidate as a representative. Attitudes toward
a candidate, therefore, are determined at two levels of perception: that
of man qua man and that of man qua symbol or representative.
C. PARTIES
The final significant electoral factor is party identification. Parties
assume particular significance in our electoral system because of our
historical experience with one party domination. Party loyalties, once
set, tend to remain set. For this reason, party identification is perhaps
the most important single)determinant of voting behavior.
Beliefs and stereotypes about parties are a major attitudinal determinant
of political behavior. In the South, Democratic loyalties remain strong
because of the force of tradition and the stereotype of the GOP as the Party
of Reconstruction (this is changing, of course, in national elections, but
not too much in state and local elections). In working class neighborhoods,
antagonism against the GOP is rooted in the belief that the Republicans
are the Party of Big Business. These are powerful attitudinal determinants,
rooted as they are in deeply held belief and well established stereotype.
A less emotional, and hence less decisive, attitudinal determinant is
the degree to which the voter identifies party with issues and interests of
importance to him. The South is the best example of this determinant at
work, for although hostility to the GOP is traditionalized by stereotype,
a perceptible shift in party loyalty has occurred as a result of voters
identifying the Republican Party nationally with those issues and interests
9
of most importance to them. Party loyalties are firm, but not fixed,
and they shift as issuc/interest identification begins to overcome beliefs
and stereotypes.
Finally, attitudes toward parties can be influenced by the suscepti-
bility of voting blocs to organizational efforts. This statement is merely
a verbal claboration of the observable practice of systematic recruiting
of people into the party through organizational techniques. Slow, tedious,
and difficult, voter attitudes toward parties can be changed through
conscious organizational efforts. However, of all the determinants of
political behavior, organizational susceptibility is probably the least
significant, at least in short-range terms.
III. PROGRAMMED POLITICAL STIMULI
To the working politician, the study of social antecedents and
attitudinal determinants of political behavior is more than an intellectual
exercise, an accounting of how and why Americans vote as they do; it
is an indespensible prerequisite for a sophisticated campaign designed to
change voting patterns by influencing political behavior.
From data setting forth the social antecedents of political behavior
it is possible to determine who votes and how they vote. Such data is
useful in identifying those groups which are most likely to support a
candidate and, once identified, special attention can be devoted to them.
10
However, what a candidate really wants to know is how hc can reach
voters not already disposed to support him, which means he wants to
know why people vote as they do and what can be done to change their
initial disposition.
Attitudinal determinants of political behavior are the key to this
vote-changing process: change attitudes and you change votes. Thus,
a well managed campaign is one in which a conscious effort is made to
alter attitudes by the application of programmed political stimuli.
If issues, candidates, and parties are the key elements in the
election process and attitudes toward these elements are central to
voting behavior, it is logical that the candidate's objective is to approach
each element with an eye on its implications for voter attitudes. Issue
development, candidate image, and party emphasis become the keys to
electoral success.
A. ISSUES
The first step is to identify which issues are important to which voting
blocs and how important they are to each. There are bound to be conflicts
among voting blocs so that it becomes necessary to make a choice as to
which bloc shall be appealed to on the basis of a particular issue. In
making this choice, it may be that an issue important to Catholics can
only be emphasized by alienating Southern Baptists; however, it may be
such alienation can be avoided by also emphasizing another issue
important to Southern Baptists but of little interest to Catholics; that
is, decisions on issue emphasis require more imput than the attitude
of contending blocs toward the single issue in question.
Some voting blocs will perceive an issue to be important which a
candidate Call not develop either because it will clearly cost him too
much with other important blocs or because the candidate is in an
inherently weak position on the issue. The best example of the latter
case was the economy during the 1970 campaign; we had to avoid the
issue because of the inherent weakness of our position. Knowing which
issues to emphasize and which issues to avoid is often as important as
knowing which issues should be developed in order to maximize strength
among a particular segment of the voting population.
Once a decision is made to develop an issue, the question is whether
it should be developed rhetorically or programmatically. This is a
major consideration for an incumbent who is in a position to deliver
programmatic solutions to issues.
Rhetoric is persuasive and profitable only when it addresses an issue
of concern to the voters in a manner that is convincing and reassuring.
Issue rhetoric must capture the aspirations of the electorate, must be
12
credible, and must hold promise of being translated into concrete
results. Rhetoric is not a substitute for programmatic development
of issues; it is a supplement, at most a holding action. Issue rhetoric
must be distinguished from image rhetoric -- the latter implicitly
involves nothing more than reassurance and confidence, while the
former necessarily implies action and resolution of issues in dispute.
Being able to determine when rhetoric is a sufficient response to the
issue-oriented concern of the electorate is an art, not a science, and
is most difficult. It is, however, also most important, particularly
when a party realignment is in progress.
If political behavior is to be altered by influencing the attitudes of
voters toward political issues, it is imperative that those issues be
defined in reference to particular voting blocs, that the attitudes be
identified, and that the response, rhetorical or programmatic, be of
such force as to have a major impact on those attitudes. Some issues
must and should be avoided, but not out of ignorance of the fact that
some voters regard them as issues. Issue avoidance should be a
calculated political decision based on an assessment that the candidate
is inherently disadvantaged on a particular issue. This disadvantage can
possibly be compensated for by development of another issue of equal
13
concern to the targeted voting bloc or by concentration on altering
that bloc's attitude toward candidate or party; the value of a systematic
analysis of the variables that influence political behavior is that it
identifies a variety of ways by which voter attitudes can be changed.
B. CANDIDATE IMAGE
No one need tell a candidate that his image before the electorate is
a decisive factor in election results. However, it does need to be
emphasized that in terms of party realignment leading to one party
domination of the political process, candidate image does not play as
important a role as issue development and party emphasis. Lincoln and
Roosevelt came and went, but the party they led remained in power long
after they had disappeared from the political scene. Candidate image
is most important in critical and deviant elections -- the former being
exemplified by the critical victories of Jefferson, Jackson, Lincoln and
Roosevelt, the latter by the deviant victories of Harrison, Taylor,
Cleveland, Wilson, and Eisenhower. Only history will tell if the Nixon
victory was critical or deviant.
Leadership (or candidacy) is generally characterized as charismatic
or institutional. The critical distinction between the two types may be
simply the manner in which the leader is perceived by the public: either
emotionally or objectively, as an unusual man whose influence is
independent of his institutional power or position, or as a more traditional
figure whose influence is dependent upon his position in or his identity
with the institutional order.
Although i! is doubtful that charisma can be artificially created, there
is reason to believe that it can be artificially enhanced; a rough diamond
can be polished, but a leopard can't change its spots (to coin a phrase).
The point, of course, is that a candidate whose appeal is essentially
institutional cannot be repackaged and marketed as a charismatic leader
without creating a credibility problem.
There is no reason to believe that charismatic leadership is inherently
superior or more successful than institutional leadership. It is essentially
a difference of style, of technique, and it is by results that history judges
the success of leadership. A non-charismatic candidate who frets about
the absence of such appeal often makes a serious mistake by attempting
to project a pseudo-charismatic image at odds with a natural image
popularly more credible and electorally more helpful. Who would have
believed General Eisenhower in Camelot?
Image, presidential and otherwise, has been over-cmphasized because
misunderstood. In a stable political order characterized by a two party
15
system, image, unless overtly offensive, is essentially a neutral
factor, less important in influencing political behavior than issues
and party. The candidate of a majority party can normally rely upon
the loyalty of his party members to carry him to victory; his only
concern is that his image not jeopardize the existing electoral fidelity
of this majority that identifies with his party. This was Lyndon Johnson's
problem. He represented the dominant political party, and in the course
of normal events, party loyalty should have been sufficient to guarantee
his re-election. However, he projected a distinctly negative image that
alienated members of his own party. Additionally (and perhaps more
importantly) his handling of the issues (race and war) alienated many
who had previously identified with his party. Lyndon Johnson didn't
need charisma to win in 1964; he would have had to have charisma plus
a bonus of good luck to win in 1968.
Candidate image is only one of three factors that influences political
behavior, and it should be considered in this limited scope. That is, it
ought to be considered in reference to the precise ways in which it can
alter political attitudes.
Affective disposition or emotion is the principal determinant of
attitudes toward a candidate and where this element is identified, steps
must be taken to deal with it. The "Tricky Dick" image, for example,
16
was such a problem. Having identified it as a problem, it was not
difficult to take positive steps designed to dispell the image.
An overlooked "image" problem relates to the manner in which key
voting blocs identify a candidate with interests they deem contrary to
their own. Thus, Lithuanians who believe a candidate is too soft on
the Soviets can be reassured by a statement to the contrary, an
appearance at an event of major interest to the Lithuanian community,
or by other gestures designed to correct the image. Likewise, union
members who believe a candidate is too closely associated with manage-
ment can be reassured by steps which are "image corrective" in nature.
Image is more than what comes across on a television screen. It is
principally a matter of perception by members of key voting blocs, and
a particular image that is favorable with one bloc may not be with
another. It is necessary therefore to identify existing image problems and
Blacks
take steps to correct them. Generally they can be corrected without
generating a new image problem with another voting bloc.
Although mass communications have tended to nationalize many social
and political attitudes, they have not eliminated all attitudinal distinctions.
As long as different political blocs with different views and attitudes exist
(that is, as long as America remains a pluralistic society), there will
be differences of image perception among the public, and no national
candidate can be satisfied with projecting a strictly "national" image.
C. PARTY EMPHASIS
In early August of 1968, it was widely believed that it was worthless
to consider past voting patterns as an indicator of future political behavior
because events were proceeding at such a rapid pace with such unexpected
results that voter attitudes were influenced only by the latest political
stimuli. At the same time, however, some skeptics were suggesting in
the face of polls showing Nixon with an overwhelming lead that by election
day traditional voting habits would begin to assert themselves, that party
loyalty would once again prove to be the most decisive (but not necessarily
the only decisive) determinant of voting behavior.
It is now generally recognized that this skeptical view was the most
practical and most accurate one. In the final weeks of the campaign
disillusioned Democrats who had toyed with the idea of voting for Wallace
or for Nixon returned to the ranks of their party. This was particularly
true of blue collar workers and others subject to the influence of organized
labor.
In spite of the rapidity of social and political change, in spite of the
obvious dissatisfaction with their party, its candidates, and its position
on the issues, large numbers of Democrats decided in the final days of the
campaign to stay with their party. A persuasive case can be made that the
18
remarkable thing about 1968 was not the number of Democrats who
bolted, but the number who stayed aboard what had all the appearances
of a sinking ship. One can and must conjecture about what would have
happened to the Wallace vote if the Governor had pulled out, but one
would be naive in the course of doing so to discount the likelihood that
traditional party loyalty would have been the decisive factor in its
distribution.
The point, of course, is that identification with party is and always
has been the principal determinant of political behavior. We identify
some elections as "deviant" because of their conspicuousness as
exceptions to the pattern of political control by a dominant party. Until
Woodrow Wilson's successful bid for re-election, no minority party was
able to elect its candidate to the Presidency for two successive terms.
Cleveland might have accomplished this feat if his electoral vote in 1888
had reflected his popular vote and he was elected to a second term in 1892.
However, it is often overlooked by casual students of this period that the
Democrats were by the mid-80s rapidly becoming the majority party; were
it not for the Depression of 1893 and the seizure of the party by the
populists in 1896, the Democrats very likely would have become the
dominant party while Franklin D. Roosevelt was still a school boy. The
election of 1896 was a critical one for the GOP.
Without taking exception to Kevin Phillips analysis, candor requires
the observation that the most discouraging aspect of the prospects for
the 1972 election is the fact that available evidence suggests that the
GOP remains, by a substantial margin, the minority party. Our
posture much more clearly parallels that of Woodrow Wilson in 1916
than it does Franklin D. Roosevelt in 1936. For this reason, the
attitude of voters toward party will be critically important in 1972 and
a fundamental and crucial decision will have to be made about the
degree to which the Republican Party will be emphasized as the vehicle
through which the aspirations of the electorate can be realized.
This will be a difficult decision because a major party alignment is
not achieved by virtue of personalities. Ike didn't do it and the
President can't do it. If a realigmment of the sort that Kevin postulates
is to take place, it must occur as the result of a conscious decision by
the voters that their interests are most clearly served by the Republican
as opposed to the Democratic Party and the President must run as a
standard bearer and not on his own. If, however, it is clear that we
are not in the process of such a realignment (or that we are not yet far
enough along in the process for it to be finally consummated in 1972),
the President's identification with the GOP may prove to be a disadvantage.
The dilemma we face is not simply one of gauging the degree of voter
identification with the Republican Party. If we assume that we are in a
1916-type situation in which party should be de-emphasized and adopt that
course, we may in the process retard what could be a significant shift in
party realignment. On the other hand, if we assume that we are in a
1936-type position in which we can move in for the kill and establish GOP
domination, we may jeopardize the President's re-election chances if we are
wrong
What is required, therefore, is some precise study and analysis of
voter attitudes toward party. To be helpful, this study and analysis must
be centered on special voting blocs in special parts of the country. Gallup
polls which purport to show the party breakdown nation-wide are not very
helpful. What we are interested in is party attitudes among critical voting
blocs whose support is imperative for a party realignment to take place.
National shifts are interesting, but not particularly important. Critical
elections occur in key geographical areas among key voting blocs; if they
shift, it won't be long before others follow.
It is not enough, of course, to identify present attitudes toward party.
What is required is to identify what can be done to change these attitudes.
This process relates almost exclusively to the development and presentation
of issues. Ike carried Michigan and Virginia; the former because of
21
personality, the latter because of issues, Only Virginia represents an
example of party realignment favorable to the GOP, and it should be
remembered that over the long-term only issue/interest identification
can permanently overcome party stereotypes -- which is to say that
candidate image may elect a President, but issues establish dominant
party control
IV. FROM THEORY TO PRACTICE
The above discussion suggests a conceptual framework for the analysis
of political behavior for the purpose of altering voting patterns. It
presupposes that the variables that determine political behavior can be
identified and, once identified, influenced. It is not a mysterious process
but merely the systemization of what every practicing politician attempts
to do intuitively. It calls for the application of modern technology and
advanced political theory to the practical problems of a Presidential
campaign.
Demographic and voting statistics are readily available and, in machine
readible form, can easily be handled by computers. From such data
sophisticated analysts can readily identify significant attitudinal determinants
of political behavior that can be influenced by programmed political stimuli.
The significant advantage of this type of program is that it provides
more precise and reliable information upon which to base sound
political decisions; it elevates the "hunch" to a fact. There is no
substitute for good political judgment - which means that the crucial
element in any process of political analysis is the analyst - but the
quality of analysis also depends upon the quality of the information
upon which it is based. Give a good analyst good information and you
are ahead of the game. We can be, if we choose to be.
TYPES OF VARIABLES RELEVANT TO THE ANALYSIS OF POLITICAL BEHAVIOR
PRESERVATION COPY
I.
Historical, Economic, Societal
Political
historical experience/heritage
party identification
cultural & religious tradition
SOCIAL
party regularity
family/peer group norms
ANTECEDENTS
process participation
class consciousness
Parties'
beliefs and
II.
stereotypes
Demographic
issue/interest
sex, age, marital
identification
status
ATTITUDINAL
organizational
race/ethnic origin
DETERMINANTS
Issues
susceptibility
religion
[
perception
Motives, interests,
IVI.
economic status
values
]
education
intensity
Candidates
place of residence
volatility
affective dis-
political charac-
terization
position
POLITICAL
objective appraisal
BEHAVIOR
party/interest
association
III.
Party Emphasis
(de-emphasis
PROGRAMMED
POLITICAL
STIMULI
Pattern of
Issue Development
Communication]
(avoidance)
Candidate Image
charismatic/institutional