=
Political
Culture and Abortion Politics in the
Br=
ian
Robert Calfano
<= o:p>
Abstract
&nb=
sp; Abortion politics has long been considered the
property of national political debate. However, the national implications of
abortion notwithstanding, this study suggests that a return to state and lo=
cal
examination of the abortion issue yields important insight into the dynamics
that local cultural variables introduce into state level policymaking.
&=
nbsp; At
the heart of this study is the question of why
(The author gratefully acknowledges the substant=
ive
and editorial contributions of Elizabeth A. Oldmixon, Ph.D., and Jason
Harrison.)
The
&=
nbsp;
&=
nbsp; This
study finds that part of the reason for
Ecclesiastical elites, in a departure from elit=
es in
other settings, derive their power and influence from a complicated balance=
between
their claims to represent divine authority and/or temporal leadership, and
their parishioners’ subsequent willingness to heed these claims. If
parishioners do not agree to confer legitimacy on the elites, or if the
parishioner community proves to be relatively uncommitted to regular contact
with elite influences, ecclesiastical elites face a diminished level of
suasion. When parishioner contact is relatively high, elite claims and
parishioner accidence become mutually affirming and dependent conditions for
elite influence over not only religious concerns, but political ones as wel=
l.
Hence, this study suggests that a palpable “pro-life” influence=
on
the tenor of local culture, emanating from certain houses of worship, resul=
ts
in predominately “pro-life” legislative action by the Pennsylva=
nia
House of Representatives and not in neighboring states in part because of
differences in the relative strength of ecclesiastical elite suasion.
Given t=
hat
that 71% of registered voters in Pennsylvania claim to favor some degree of
restriction on abortion (ranging from parental notification to a ban on
late-term abortions), and 24% favor the outright banning of reproductive
choices entirely, these conservative policies should perhaps not be surpris=
ing
(Quinnipiac University Poll, June 5, 2002). However, despite the strong
conservative indicators among the
<=
span
style=3D'font-family:Arial'>The State of the Party
<=
span
style=3D'font-family:Arial'>
<=
span
style=3D'font-family:Arial'>
<=
span
style=3D'font-family:Arial'>Table 1: Voting Position on Reproductive Rights
According to Party—PA, NY, NJ
2002
<=
b>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total Number of
Representatives |
150 (139 Included in
Study) |
|
Democrats |
100 |
|
Republicans |
50 |
|
“Pro-lifeR=
21;
Democrats |
5 |
|
“Pro-lifeR=
21;
Republicans |
32 |
|
Total Number of=
Representatives |
80 (67 Included=
in
Study) |
|
Democrats |
47 |
|
Republicans |
33 |
|
“Pro-life=
”
Democrats |
1 (1 Abstention=
) |
|
“Pro-life=
”
Republicans |
29 (6 Abstentio=
ns) |
(Sources:
Elites & Political Culture If party
identification alone is not able to explain why Pennsylvania Democrats appe=
ar
evenly divided over reproductive rights, perhaps other sources of influence=
are
at work affecting representative position either directly and/or indirectly.
Such organized sources of influence, in order to be salient in districts, w=
ill
likely be comprised of local voting blocs of varying degrees of size and
sophistication. Though
Zaller’s contention regarding opinion fluctuation among the public is=
valid,
this study suggests that public opinion, when broken down along
organizational lines, and influenced by the referent, legitimate, and/or ex=
pert
powers of elite figures of those organizations, functions as an
effective causal influence over the local political culture, and subsequent=
ly,
a representative’s political positions (Pillari and Newsome, Jr.,
1998, 109). In other words, elite communities—those persons responsib=
le
for providing leadership in local organizations—serve as aggregation
points and cueing agents for the voting public. These aggregation points, if
they are well organized and send consistent messages, become influential no=
des
in a legislative district. Regardi=
ng
reproductive rights policies, few elite groups possess the degree of local =
salience
and credibility to function as influential nodes as do those of organized
religious communities—parishes and congregations (Verbit, 1970, 36-37=
).
Hence, those with administrative and spiritual responsibility over these
communities, what are referred to here as ecclesiastical elites,
function as a causal explanator (in conjunction with party identification,
race, education, and other controls) of the variation in abortion position
among representatives in this three state study. Table 2
delineates the relative size of the evangelical Protestant and Roman Cathol=
ic
communities in the three states considered. These figures, measuring the nu=
mber
of parishes and congregations, serve as proxy measurement for the size of t=
he
ecclesiastical communities in each state.&=
nbsp;
Table 2: Comparative
Christianity—PA, NY, and NJ 2000 Total Number of Faith Communities 12,512 Roman Catholic Parishes 1,624 Evangelical Protestant Parishes 3,906 Total Number of Faith Communities 10,203 Roman Catholic Parishes 1,784 Evangelical Protestant Parishes 2,750 Total Number of Faith Communities 4,214 Roman Catholic Parishes 774 Evangelical Protestant Parishes 1,070 (Source: American Religion Data
Archive, 2000) As
organizations possessing a distinct hierarchy of elites and considerable
infrastructure, district parishes and congregations have the potential, and,
perhaps most importantly, the motive, to influence representative
abortion position (Cook, Jelen, and Wilcox, 1993, 771-781). Hence,
ecclesiastical elites, and members of elite communities in general, =
who
take a strong, consistent, and public stand on political and/or cultural
issues, can influence the local political culture toward a tenor that is in
proximity to their preferences (Kelly and Duerst-Lahti, 1995). These e=
lites
possess cultural and political influence because of the role that they play=
in
the lives of their organizations’ membership. As leaders of faith
communities in particular, ecclesiastical elites function as the conduits of
emotional comfort and spiritual assurance. They further function as
interpreters of human mysteries, especially for those things that inspire a=
we
and create a sense of dependence and human connectivity (Zeitlin, 2004, 28-=
31;
173). It is this close connection that ecclesiastical elites possess with t=
heir
faith communities in matters of the emotional and spiritual that parlays in=
to
influence on political and cultural issues. In maki=
ng the
connection between elites and cultural/political influence, Oldmixon sugges=
ts
that “. . . decision making on reproductive rights is strongly influe=
nced
by an array of cultural indicators. Elite-level cultural considerations, in
particular, seem to affect decision making. . . . culture theory is a useful
theoretical lens through which to view the institutional politics of
reproductive policy” (2002, 776). Concomitantly, because,
demographically, traditionally strong supporters of the “pro-lifeR=
21;
view “are not particularly well endowed with participatory
resources,” ecclesiastical elites have an even greater organizational
role to play than they might in other political and cultural concerns (Verb=
a,
1996, 5). This st=
udy
proceeds from the view of culture as the amalgam of local elite and
organizational influences. It is these competing elite influences that seek=
the
attention of, and influence over, district representatives in order to addr=
ess
political concerns. Always in a potential state of flux, the tenor of local
political culture can differ from locale to locale (district to district in
this case) depending on the comparative strength of the respective elite
communities on both sides of an issue, abortion rights in this case. As suc=
h,
there can be variation in that which is considered morally and politically
acceptable among various locales within and between states. It is
necessary at this juncture to develop a satisfactory working understanding =
of
culture. For the purposes of this study, culture is the product of societal
consensus regarding the boundaries of acceptable behavior. More than a set =
of
rules and regulations, however, culture represents a fluctuating amalgam of
elite cross pressures, political ideologies, party agendas, and disparate
concepts of epistemology and human development along gender, race, and
socio-economic lines (Gilligan, 1982, 171-174; Rothenberg, 1995).
Concomitantly, culture represents a set of symbols, norms, and values that
function as civic touchstones (Laitin, 1986, 23-50). Though =
this
study refers to commonly accepted norms as simply culture, it is
inescapable that the political process, especially in a stable democratic
republic, is a primary vehicle for addressing many social concerns, especia=
lly
reproductive rights issues (Leege, Wald, Krueger, and Mueller, 2002, 26-27).
Hence, it is necessary to recognize the truly political context of the issu=
es
over which elites, ecclesiastical and otherwise, possess a degree of influe=
nce.
On both sides of the abortion debate, elites capitalize on their positions =
of
influence to induce followers to adopt a particular life view and lexicon
(Chilton, 1988, 419-455). Through elite level encouragement, group members
develop a life view that includes a political dimension. This political
dimension includes a schema that is in harmony with the value-laden cues
received from elites. It is when political representatives become aware that
elites command influence over important constituencies within their distric=
ts
that elite influence take shape. The result of this elite influence, and
division, over the control of the cultural tenor, has been described as
“culture war” by some observers (Duerst-Lahti and Verstegen, 22=
5,
1995). The Re-Elective Motive The pri=
mary
theoretical backdrop for the understanding how elites, ecclesiastical and
otherwise, might influence local culture and representative decision is
Mayhew’s seminal thesis regarding political actors and their primary
vocational concern—re-election (1974, 44-47). Although Mayhew conside=
rs
members of the U.S. Congress in his study, the theory can be transferred to=
the
state district level. According to Mayhew, state representatives will adopt=
an
abortion position that is congruent with those of prominent local elites who
possess a palpable degree of constituent influence. Implicit in this argume=
nt
is that, if elites are ignored, or such a perception exists, they will use
their influence with rank and file to exact a measure of electoral punishme=
nt.
To have such influence, Olson suggests that organizational membership size =
must
be relatively conducive to political mobilization—not overwhelmingly
large (Olson, 1965, 165). This is=
not
to suggest that a representative cannot, and does not, develop her issue
positions distinct from those of local elites. Yet, in keeping with
Mayhew’s theory, and the earlier findings of Miller and Stokes, she w=
ill
do so only when there is minimal political risk involved; in other words, w=
hen
re-election is not at stake (1963, 56). Risks involving alienation of a lar=
ge
number of constituents, on both sides of the issue, increase when reproduct=
ive
rights are involved. Part of the reason for this condition is the comparati=
ve
degree of constituent comprehension of what is at stake. Since abortion and
other social issues are relatively easy to understand (in contrast with bud=
get
and taxation) many constituents are likely to base their votes on such issu=
es,
especially with elite encouragement (Meier and Haider-Markel 332-334,
1996). In the
process of catering to those political and community elites whose influence
appears critical for re-election, there will be those elites whose influence
over policy development is diminished (Aldrich and Rohde, 274, 2001). That =
some
elites experience shifting degrees of influence in relation to others appea=
rs
especially true when considering the factional struggle over establishing a=
nd
maintaining the tenor of a local political culture. As such, especially with
controversial social issues, it might be possible for representatives to
compensate for alienating certain elite communities by acceding to the dema=
nds
of others. This might help explain the different positions taken on
reproductive rights among members of the state legislature. Previous Scholarship and Approaches Interes=
tingly,
the extant literature has treated reproductive rights as a mostly national
concern. As such, existing studies employ national voting data to tap
indicators of abortion support or opposition among the electorate. (Legge,
1983, 481-486). Those who have examined the issue at the state level either
have approached state-specific developments as outgrowths of the federal
abortion debate, or have aggregated state-level data into studies within the
national framework (Roh and Haider-Markel, 2003, 17-19). While it is likely
that interest groups and elites in the federal abortion debate play an
important role in organizing and equipping their local affiliates, a broad
empirical approach to state-level politics will likely miss specific local
elite and cultural dynamics (Lowery and Sigelman, 1982, 383). Existing
literature also tends to minimize the role of culture as an indicator of po=
licy
development and representative positions. There are, however, some theoreti=
cal
threads of the cultural thesis to be seen. These threads suggest that “variation in state-lev=
el
partisanship and ideology is due to state-level differences in political
culture rather than the demographic characteristics . . . it is clear that =
the
political attitudes of Culture as Motive Certain=
ly, a
valid critique of the culture-based approach is to suggest that culture, as
influenced by parishes and congregations on reproductive rights issues in
specific, is a misspecification of what is actually driving representative
abortion positions—generally conservative political preferences withi=
n a
district, state, and/or region. While this is a valid critique, such an
explanation leaves to be seen how and why districts will possess
conservative preferences, especially on social policy issues. In other word=
s,
there must be something driving these conservative preferences that cannot =
be
assumed to exist prima facie. Part of
understanding a district’s conservative preferences comes through
distinguishing between different types of conservative ideology. There is a
difference between economic conservatism, which is concerned almost exclusi=
vely
with limiting government activities in particular policy realms, and social
conservatism, which attempts to limit the scope of acceptable behavior for
individuals, including, but not limited to, abortion. Hence, a district tho=
ught
to be generally conservative on fiscal policy is not necessarily explaining
much about the locale’s socially conservative tenor, to the de=
gree
that it has one. What=
217;s
more, cultural wars center more frequently on those social issues that tend=
to
elicit strong conservative and liberal responses because such issues have
become salient in the minds of elites and constituencies. It is suggested,
therefore, that cultural dynamics, rather than any intrinsic set of
characteristics, contributes in making a district conservative (or not) on =
the
abortion issue. When the issue involves abortion, Reichley suggests a strong
influence for ecclesiastical elites. “Religion has been the underlying
thread of moral unity . . . the main threads of our religious tradition have
fostered a shared commitment to . . . the sanctity of human life . . .̶=
1;
(1986, 47). Ecclesi=
astical
elites are not necessarily strong influences on culture and representative
position when the issues are of an economic concern specifically. This is w=
hy
party organizations, civic groups, neighborhoods, the business community,
artistic guilds, and non-profit social service providers all play an import=
ant
role in creating, maintaining, and changing local culture. Depending upon t=
he
array of issues that are salient in a district, the ecclesiastical elites m=
ight
actually not be considered at all influential in the local cultural calculu=
s.
Hence, any single variable distinction (i.e., party identification or elite
influence over all other variables) that some might make concerning
representative position taking on abortion seems to be fundamentally
wrong-headed. Instead, it is far more fruitful to treat a panoply of casual
variables as equally important indicators of elite influence and cultural t=
enor
in a legislative district. A Religious Divide Christi=
an
organizations and ecclesiastical elites have been, until this point, treate=
d as
a homogenous group. It would be fallacious, however, to assume that all
ecclesiastical elites and Christian organizations seek to instill a
“pro-life” influence on the local culture and state
representatives. For instance, that much of the leadership, and a large
minority of the congregants, of mainline Protestant denominations have adop=
ted
a “pro-choice” position is well documented. Yet the=
re is
reason to suspect that the even in the evangelical wing of
Protestantism, either because of an aversion to political involvement in so=
me
quarters, or actual disunity over political ideology, there is dissention on
abortion (Watt, 2002, 109; Hart, 2002, 169-171). With only 5% of evangelica=
ls
claiming to be “pro-choice” in recent polling, this is not to
suggest an evangelical sea change on abortion (Barna Research Group, 2001).
What some studies do suggest, however, is that the evangelical Protestant e=
lite
structure at the local level is not homogenous enough, or requisitely
organized, to speak with clarity on abortion issues (Dayton and Johnson, 19=
91).
Part of=
the
difficulty confronting the evangelical community in this regard is its hist=
ory
as a displaced social force that preferred to concentrate on spiritual matt=
ers
rather than “‘secular’ politics” (Wald, 2003, 200-2=
08).
While evangelicals have organized for national political efforts in recent
decades, there remains an intellectual tension among evangelical elites
regarding the degree to which Christianity=
and its institutions should be used for political and cultural influ=
ence
(Moltmann, 1986). Thus, while there are undoubtedly those evangelical elites
who are willing and eager to attempt to influence local culture, there are =
also
those who are not so primed for the activity. Concomi=
tantly,
evangelical Protestants appear divided over more than cultural and political
questions; theology is also a divisive issue. Recent polling finds that, am=
ong
evangelical pastors, 71% of Southern Baptist, 51% of non-denominational, 44=
% of
charismatic, and 35% in black congregations hold to what the Barna Research
Group calls “a biblical worldview” (2001). The Barna definition=
of
this “worldview,” with which evangelical denominations have been
traditionally associated, espouses a belief in the accuracy of the biblical
text, the sinless earthly existence of Jesus Christ, and the doctrine of
salvation through spiritual grace alone. That a large number of evangelical
elites cannot agree on these doctrinal essentials, raises serious questions
concerning the community’s homogeneity. =
Regarding
evangelical elite cleavage on reproductive rights, Fowler documents an
evangelical community that, post Roe v. Wade, possesses a number of
dissenting voices in elite positions who advocate a limited
“pro-choice” public policy (1987, 77-91). Though a majority of
evangelicals claim to be “pro-life,” it is the ecclesiastical
elites who have the primary role of transforming the opinion inconsistencies
that Zaller elucidates into a focused cultural influence. That there appear=
s to
be differences on elite abortion position suggests that evangelical Protest=
ant
congregations, perhaps much like their mainline counterparts, have an
inconsistent influence on furthering the “pro-life” position in
local culture, especially under statistical examination (Huckfeldt, Plutzer,
and Sprague, 1993, 277-379; Huckfeldt et al. 1995, 1044). By cont=
rast,
studies suggest that Roman Catholic elites are fairly influential in shaping
their parishioners’ political views, especially among regular parish
attendees. Interestingly, Catholic elites are able to do so while avoiding a
clear voting allegiance to either major political party (Steinfels, 2003, 8=
0).
Much of the reason for elite consistency is attributable to Roman
Catholicism’s hierarchical authority structure (McMullen, 1994, 712;
Burns, 1992, 719-720). This structure contributes to the generally consiste=
nt
respect that local elites show to papal decrees that label abortion and
infanticide as “abominable crimes” (Charles, 1999, 21). Though there have been efforts at
laity-centered reform within Catholicism, many of these projects have been
either quelled at the episcopal level, or have not posed significant challe=
nges
to the church’s abortion position (Roche, 1970). Catholic
elites are able to use the common symbolic and value framework that they cr=
eate
in the parish community, in conjunction with an abortion position that cont=
ains
minimal variance, to harness, or at least give the appearance of harnessing,
the collective political action and opinion of their parishioner communities
(Ammerman, 2001, 362-370). This harnessing signals the district representat=
ive
regarding the potential for electoral punishment for going against the Cath=
olic
position on the issue. Importa=
ntly,
church decrees against political partisanship, far from being a hindrance to
political influence, affords Catholic elites an audience with those local
political elites who might be suspicious of the partisan reputation of some=
in
the evangelical elite community (Willis, 1990, 242). (An important caveat to
understanding elite influence within Roman Catholicism is that, to be effective, there must be a sizeable
percentage of parishioners who place themselves in regular proximity to
ecclesiastical elite influence. Without a sizeable cadre of parishioners to
influence, elite input will be severely limited in the cultural-political
context. ) Evangel=
ical
elites, because they are scattering along various denominational, and even
non-corporate boundaries, find it more difficult to establish cultural
influence. No single evangelical denomination, or group of elites, can speak
for a comparatively sized proportion of the Christian community as is the c=
ase
with the Roman Catholic Church. At the same time, American evangelicalism
struggles with racial cleavages that divide it politically, historically, d=
octrinally,
and culturally (Cone, 1986). A
commensurate level of cultural difference does not exist in Catholicism, at
least with parishioners who regularly attend Modeling Local Culture As stat=
ed,
local culture is the amalgam of a disparate set of organized actors, of whi=
ch
the influence of ecclesiastical elites is but one component. Hence, the
logistic regression models below include controls for differences in popula=
tion
centers among the three states, representative party affiliation, education
levels, the percentage of African-American residents per district, and per
capita income per district. Though the components of local culture certainly
encompass more than the variables employed here, it is believed that the mo=
st
important variables are represented. (Implicit in these models is the
assumption that the number of parishes and congregations in each district a=
re
adequate proxy representations of the size of the local ecclesiastical elite
communities.) 1) Y (Member’s Abortion Position) =3D B0 + B1 (Proportion of Roman
Catholic Parishes in a PA House District) + B2 ( 2) Y (Member’s Abortion Position) =3D B0 + B1 (Proportion of
Evangelical Protestant Congregations in a PA House District) + B2 (<=
st1:place> These models are run in the same format for the <=
/b> &=
nbsp; Hypothesis
One (Model 1): There are organizational and polit=
ical
differences between Roman Catholic parishes and Evangelical Protestant
congregations that might make the former’s elite community more influential as a force of loc=
al
cultural influence with regard to reproductive rights. Hence, with Roman Ca=
tholic
parishes isolated, it is expected that as the proportion of Roman Cathol=
ic
parishes, and, consequently, the number of Catholic elites, increases in
a House district, so does the likelihood that the district representative w=
ill
adopt a “pro-life” position. &=
nbsp; Hypothesis
Seven (Income Control): Hall and Ferree also find a direct a=
nd
positive relationship between income level and the “pro-choice”
position (201, 1986). It is expected, therefore, that districts with hig=
her
per capita income levels will have members with “pro-choice”
positions. Measuring
Representative Abortion Position The
representative’s identification as either “pro-life” or
“pro-choice” is this model’s dependent variable. As no
existing data set taps abortion positions for all state legislative distric=
ts
in the three states examined here, an original data set, which codes each H=
ouse
member in Those
representatives designated as “leaning” toward either abortion
position were coded as having that particular position. Those representativ=
es
for whom information was not provided by the afore-mentioned organizations =
were
contacted directly regarding their position. Those representatives unwillin=
g or
unable to provide information on their abortion position in any of the three
states have been omitted. Culling
the Congregations and Parishes The model’s main causal variable is the
proportion of conservative parishes and congregations within each of the 203
Pennsylvania House districts, the 150 districts in One could suggest that grouping all conservative
parishes and/or congregations under a “pro-life” banner is
committing an ecological fallacy. This suggestion would be warranted if the
design were measuring the proportion of adherents per district. Agai=
n,
while adherents have an important role to play in the process of influencing
representatives, it is the cues that both adherents and representatives rec=
eive
from ecclesiastical elites that is suggested here to be the more accurate
measurement of ecclesiastical influence on the local culture. The decision to use the ARDA’s
“evangelical Protestant” label, without more specific distincti=
ons
between fundamentalists, charismatics, and evangelicals, might encounter so=
me
objection, as there is some variance of evangelical views on abortion betwe=
en
elites in these sub-categories
(Smidt, 612, 1988). This is certainly a relevant critique of the ove=
rall
design method. However, given that the existing literature does not couple
specific denominations with corresponding para-church subsets of Protestant
Christianity—fundamentalism, evangelicalism, etc.—assigning
denominations and elites to paricular sub-categories would prove exceedingly
difficult, if not empirically misguided. Another possible objection relates to the exclu=
sion
of all mainline Protestant congregations from the count. While there is
theoretical justification for the mainline exclusion, as much of the mainli=
ne
elite structure, especially in the upper echelons, holds the “pro-cho=
ice”
position, there remain a moderately large number of local elites that hold =
to a
“pro-life” view. This appears to be the case especially in The ARDA’s parish/congregation count excl=
udes
many houses of worship that are unaffiliated with denominational organizati=
ons,
but that might be members of ecumenical federations, or, on the other hand,=
are
independent of all relationships with larger organizations (Lindner, 129,
2003). Hence, with these missing congregations not included, the proportion=
of
evangelical Protestant congregations, and elites, might be both inflated and
deflated as the circumstances dictate. The exclusion of these houses of wor=
ship
is an acknowledged difficulty with the current data set. A final aspect of the parish/congregations codi=
ng
procedure requiring explanation is the transformation of the ARDA data (whi=
ch
is broken down according to county) into the legislative district proportion
percentages. Approximately one-third of Based on the proportion of votes cast from each
county in the district, a percentage of county votes from the district vote
total were calculated. This county vote percentage was then applied to the =
ARDA
county data in order to specify the proportion of conservative county paris=
hes
and congregations that should be included for each multi-county district. T=
he
proportion of conservative parishes and congregations per county (which was
then apportioned to the districts as necessary) was calculated by aggregati=
ng
the number of houses of worship affiliated with the evangelical Protestant =
and
Roman Catholic traditions, and comparing these figures with the total numbe=
r of
ARDA parishes or congregations in each county. Controlling
for Cultural Variables Five control
variables are used in this model. The first is for population centers
designated as urban by the US Census. Represe=
ntative
party identification serves as the second control variable. Existing studie=
s on
reproductive rights and political parties suggest that party does matter in
terms of member abortion position—especially in the sense that Republ=
ican
office holders tend to be “pro-life” (Adams, 723-727, 1997). The
data presented in table one bears this reality out in the 2000 US
Census data provides the necessary information for coding the remaining con=
trol
variables. The proportion of African-American residents per state district =
is
represented as a ratio-level figure, as is the percentage of residents hold=
ing
a bachelor’s degree or higher. The district’s average per capita
income level, is coded a “1” if it is over the state’s
average as indicated by US Census data, and “=
0”
if it is not. Tables 3, 4, and 5 delineate the state-by-state results from =
the
robust logistical regression tests. Table 3: Roman Catholic and Evangelical
Protestant Parishes & Representative Position on Reproductive Rights
Legislation in the Unstandardized Logistical Regression Coefficien=
ts Independent Variables Coefficient (Model 1) Robust SE (Model 1) Coefficient (Model 2) Robust SE (Model 2) RC (1) EP (2) .0946439** .0415061 -.0577928** .0243744 Urban -.2347115 .7891403 .0510054 .7267018 Party 2.423032*** .5090397 2.220103*** .521349 Education -.470503 .0415882 -.0350031 .0452689 Race -.0618583*** .236612 -.0690026*** .0218227 Income -.9822423 .6566254 -1.63167** ,7321637 Constant 1.120107 1.036401 4.40613 1.133275 =
<=
span
style=3D'mso-spacerun:yes'> =
&nb=
sp; =
&nb=
sp; =
=
Wald Chi2 =
41.60 =
&nb=
sp; =
&nb=
sp;
36.12 =
Log-Likelihood
-77.163392 &=
nbsp; &nbs=
p; &=
nbsp; &nbs=
p;
-77.21378 =
Pseudo R2 =
0.2981 =
&nb=
sp; =
&nb=
sp;
0.2976 Predicted Probabilities .49 (-1Std. Dev.) .70 (+1 Std. Dev.) / .69 (-1 Std. Dev.) .75 (+1 Std. Dev.) =
&nb=
sp; =
&nb=
sp; =
&nb=
sp; =
&nb=
sp; =
&nb=
sp; Table 4: Unstandardized Logistical Regression Coefficien=
ts <=
span
style=3D'font-family:Arial'>Independent <=
span
style=3D'font-family:Arial'>Variables <=
span
style=3D'font-family:Arial'>Coefficient <=
span
style=3D'font-family:Arial'>(Model 1) <=
span
style=3D'font-family:Arial'>Robust SE <=
span
style=3D'font-family:Arial'>(Model 1) <=
span
style=3D'font-family:Arial'>Coefficient <=
span
style=3D'font-family:Arial'>(Model 2) <=
span
style=3D'font-family:Arial'>Robust SE <=
span
style=3D'font-family:Arial'>(Model 2) RC (1)
EP (2) .1115215 .1253758 .2207194 .1518513 Urban 2.517509* 1.39833 2.685197 1.325646 Party 3.832596*** .7733212 4.009447*** .7917291 Education -.0613125 .0572834 -.021711 .0735156 Race -.1526557 .0709014 -.304024* .1592624 Income .2243185 1.03703 .2371464 1.130225 Constant 1.120107 1.036401 -6.741215 3.827357 Wald Chi2 =
45.75 =
&nb=
sp; =
&nb=
sp;
43.31 &=
nbsp; =
Log-likelihood =
-41.030087 =
&nb=
sp;
=
-39.99388 =
Pseudo R2 =
0.4840 =
&nb=
sp; =
&nb=
sp;
0.4970 TABLE 5: Roman Catholic and Evangelical Protestant Paris=
hes
& Representative Position on Reproductive Rights Legislation in =
the Unstandardized
Logistical Regression Coefficients Independent Variables Coefficient (Model
1) Robust
SE (Model
1) Coefficient
(Model
2) Robust
SE (Model
2) RC
(1) EP (2) -.0632552 .3197745 -.573242** .237671 Urban -.2583261 1.293973 .1221182 .204272 Party 6.699595*** 1.5485 8.867328*** 1.75695 Education -.1168281 .1363114 -.3343359 .0735156 Race -.0347883 .0472893 .0675969 .1054756 Income 3.74126* 1.94393 2.38911*** .8157815 Constant 1.051841 2.565779 17.7637 6.81131 &nbs=
p;
=
Table 3 describes the results from the As predicted, representative party identificati=
on is
a highly significant variable in
=
; Hyp=
othesis
Six (Race Control): Ransford
and Miller find that differences in race among district elites can lead to
differences in abortion attitudes, especially as it regards an increasing p=
ercentage
of African-American residents, who are generally more supportive of
reproductive rights (46-51, 1983). Therefore, it is expected that a significant a=
nd
negative relationship between districts with larger percentages of
African-American residents and the “pro-life” position among st=
ate
representatives will exist.
Roman Catholic and Evangelical Protestant Paris=
hes
& Representative Position on Reproductive Rights Legislation in the <=
st1:place>
=
Wald Chi2 =
51.99 =
&nb=
sp; =
&nb=
sp;
60.57
Log-likeli=
hood =
-8.2697677 =
&nb=
sp; =
-6.3622033 =
Pseudo R2<=
span
style=3D'mso-spacerun:yes'> =
0.8205 =
&nb=
sp; =
&nb=
sp;
.08619
=
Predicted Probabilities =
.82
(-1 Std. Dev.) .8=
7 (+1
Std. Dev.)
Parish
Politicking, Party, and Culture
B=
ased on
these results, it is possible to conclude that party, race, income, and =
the
proportion of Catholic parishes and elites, function as significant,
though, as the coefficients show, not equally sized, pieces of the <=
st1:State>
The pre=
dicted
probability that a Pennsylvania representative from a district where the
proportion of Catholic parishes, given the variable specifications made in =
the
model, and holding all other variables at their means, increases is .49 for=
one
standard deviation below the mean, and is .70 for one standard deviation ab=
ove.
This finding indicates that there is a .21 increase in the probability that=
a
Catholic parishes’ (elites) legislators’ abortion position will=
be
“pro-life.” By contrast, the predicted probability for evangeli=
cal
Protestant congregations’ (elites), given the model specifications, a=
nd holding all other variables at their means, is .69 fo=
r one
standard deviation below the mean, .75 for one standard deviation above. He=
nce,
there is a .06 decrease in the probability that the evangelical Protestant
congregations’ (elites) legislators’ position will be
“pro-life.”
That the
proportion of Catholic parishes per legislative district was not significan=
t,
and, in some cases, negative, in the
Table 6: Comparative Attendance Percentages (Roman Catho=
lic
Dioceses PA, NY, NJ)
|
|
41.0% |
|
|
25.6 |
|
|
20.4 |
|
|
30.3 |
|
|
36.3 |
(Source: Hadaway, Marler, and
Chavez, 746, 1993)
Regardi=
ng the
results for the evangelical Protestant variable, which are negative and
significant for
For ins=
tance,
it might be that, at least in
This is=
a
good, though not completely satisfying, alternative interpretation.. For
example, it is not clear that districts with high congregational proportions
automatically have “pro-life,” Republican members. The mean
proportion of congregations for Republican districts in
Conclusions
At the =
heart
of this study is the question of whether the abortion position taken by
representatives in the state legislatures in
The important discovery made here i=
s that
Roman Catholic ecclesiastical elites, as represented by the proportion of
parishes in each legislative district, have a positive and significant effe=
ct
on representative abortion position in local culture in the
As stat=
ed
previously, it is likely that the Roman Catholic Church’s strong
hierarchical structure, which empowers its elites to speak for a larger and
more organized segment of the potential local voting population contributes=
to
its influential power. This, coupled with the mostly consistent
“pro-life” stand taken by the majority of the Roman Catholic
hierarchy, makes Roman Catholic parishes and elites a truly potent, though,
again, not the only, influence on local culture.
This is=
also
not to suggest that there are not ways in which evangelical elites can have=
a
similar influence as their Catholic counterparts. To the extent that local
evangelical elites hold wide community recognition, or form ad hoc &=
#8220;pro-life”
coalitions with parish elites, they might also be considered influential
factors in member abortion position selection. Part of the difficulty facing
the evangelical elites, however, is that their relative political and
theological disunity does not readily lend itself to robust statistical
causality. This might be how and where case-level studies of political
influence and ecclesiastical elite behavior, a la Wood’s recent resea=
rch,
would be a useful approach to studying local culture and religious variable=
s in
a deeper contextual setting (326-328, 1999).
In what=
ever
direction future research agendas move, it is important to keep the finding=
s of
this design in mind—local culture can be influenced significantly not
only by those variables long associated with political position-taking, but=
by
ecclesiastical elite behavior as well. Such elite behavior may, in fact, ma=
ke
the difference between states that tend to conform to existing theory on so=
cial
and political policy developments, and those, like Pennsylvania, that prese=
nt
themselves as true puzzles.
References
Ada=
ms, Greg
D. 1997. “Abortion: Evidence of an Issue Evolution,” America=
n
Journal of Political Science 41 (July): 718-737.
Ald=
rich,
John H. 1995. Why Parties? The Origin and Transformation of Political
Parties in
Ald=
rich,
John H., and David W. Rohde. 2001. “The Logic of Conditional Party
Government: Revisiting the Electoral Connection.” =
In Congress
Reconsidered, 7th ed.
Ame=
rican
Anglican Council. 2004. Member parishes.
http://www.americananglican.org(
Ame=
rican
Religious Data Archive Statistics, 2000. http://www.thearda.com
(
Amm=
erman,
Nancy Tatom. 2001. Congregation and Community.
Amm=
erman,
Nancy T., Jackson W. Carroll, Carl S. Dudley, and William McKinney.
1998. S=
tudying
Congregations: A New Handbook.
Arc=
eneaux,
Kevin. 2002. “Direct Democracy and the Link Between Public Opinion
and State Abortion Policy,” State Politics and
Policy Quarterly 4 (Winter): 372-388.
Bar=
na
Research Group, Ltd. 2001. “Born Again Adults Remain Firm in Oppositi=
on
to
Abortion and Gay Marriage.” July 23.
http://www.barna.org (
Bar=
na
Research Group, Ltd. 2004. “Only Half of Protestant Pastors Have a Bi=
blical
Worldview.”
January 12. http://www.barna.org (
Bur= ns, Gene. 1992. “Commitments and Non-Commitments: The Social Radicalism <= o:p>
of
Byr=
nes,
Timothy A. 1993. “The Politics of the American Catholic Hierarchy,=
221;
Political
=
&=
nbsp; Science
Quarterly 108
(Autumn): 497-514.
Cha=
rles,
Rodger. 1999. An Introduction to Catholic Social Teaching.
Ignatius Press.
Chi=
lton,
Stephen. 1988. “Defining Political Culture,” The Western
Political Quarterly
=
&=
nbsp; 41 (September): 419-455.
Con=
e, James
H. 1986. “Black Theology as Public Theology in
Religion and Political Theology, ed. Leroy S. Rouner. Notre D=
ame:
Coo=
k,
Elizabeth Adell, Ted G. Jelen, and
Cultures and Public Opinion about Abortion,” Po=
litical
Research Quarterly 46 (December): 771-781.
Evangelicalism. <=
st1:City>
The ‘Year of the Woman’ and Women’s
Political Representation.” In Gender Power, Leadership, and Govern=
ance,
ed. Georgia Duerst-Lahti and Rita Mae Kelly.
Eri=
kson,
Robert S., John P. McIver, and Gerald C. Wright, Jr. 1987. “State
Political
Culture and Public Opinion,” The American Polit=
ical
Science Review 81 (September): 797-814.
Fow=
ler,
Paul B. 1987. Abortion: Toward an Evangelical Consensus.
Multnomah Press.
Fre=
eman,
Jo. 1986. “The Political Culture of the Democratic and Republican
Parties,” Political Science Quarterly 101:
327-356.
Gil=
ligan,
Carol. 1982 [1993]. In a Different Voice: Psychological Theory and
Women’s Development.
Gre=
en, John
C., and James L. Guth. 1991. “Religion, Representatives, and Roll
Calls,” Legislative Studies Quarterly 16
(November): 571-584.
Gut=
h, James
L., and John. C. Green. 1990. “Politics in a New Key: Religiosity and=
Participation among Political Activists,” The
Western Political Quarterly
43 (March): 153-179.&nbs=
p;
Hai=
der-Markel,
Donald P., and Kenneth J. Meier. 1996. “The Politics of Gay and
Lesbian Rights: Expanding the Scope of the Conflict,R=
21; The
Journal of Politics 58 (May): 332-349.
Hal=
l,
Elaine J., and
Attitudes,” Public Opinion Quarterly 50
(Summer): 193-207.
Har=
t, D.G.
2002. That Old-Time Religion in Modern
Protestantism in the Twentieth Century:
Hei=
neman,
Kenneth J. 1998. God is a Conservative: Religion, Politics, and Morality=
in Contemporary
Him=
melstein,
Jerome L., and James A. McRae, Jr. 1988. “Social Issues and
Socioeconomic Status,” Public Opinion Quarterly=
52
(Winter): 492-512.
Huc=
kfeldt,
Robert, Paul Allen Beck, Russell J. Dalton, and Jeffrey Levine, 1995.
“Political Environments, Cohesive Social Groups, a=
nd
the Communication
of Public Opinion,” American journal of Politic=
al
Science 39 (November):
1025-1054.
Huc=
kfeldt,
Robert, Eric Plutzer, and John Sprague. 1993. “Alternative Contexts o=
f
Political Behavior: Churches, Neighborhoods, and
Individuals,” The
=
&=
nbsp; Journal
of Politics 55
(May): 365-381.
Kel= ly, Rita Mae, and Georgia Duerst-Lahti. 1995. “The Study of Gender Power and <= o:p>
Its Link to Governance and Leadership.” In Gend=
er
Power, Leadership, and Governance, ed. Georgia Duerst-Lahti and Rita Mae
Kelly.
Kel=
ley,
Jonathan, M.D.R. Evans, and Bruce Headey. 1993. “Moral Reasoning and =
Political Conflict: The Abortion Controversy,” =
The
British Journal of Sociology 44 (December): 589-612.
Lai=
tin,
David. 1986.Hegemony and Culture: Politics and Religious Change among
the Yoruba.
Lee=
ge,
David C., Kenneth D. Wald, Brian S. Krueger, and Paul D. Mueller. 2002.
The Politics of Cultural Differences: Social Change in t=
he
Post-New Deal Period.
Leg=
ge,
Jerome S., Jr. 1983. “The Determinants of Attitudes toward Abortion in
the
American Electorate,” The Western Political
Quarterly 36 (September): 479-490.
Lin=
dner,
Eileen, W., ed. 2003. Yearbook of American and
Low=
ery,
David, and Lee Singleman. 1982. “Political Culture and State Public
Policy:
The Missing Link,” The Western Political Quarte=
rly 35
(September): 376-384.
May=
hew,
David R. 1974. Congress: The Electoral Connection.
University Press.
McM=
ullen,
Mike. 1994. “Religious Polities as Institutions,” Social For=
ces 73
(December): 709-728.
Old=
mixon,
Elizabeth A. 2002. “Culture Wars in the Congressional Theater: How th=
e
US House of Representatives Legislates Morality,
1993-1998,” Social Science Quarterly 83 (September):
775-788.
Ols=
on,
Mancur. 1965. The Logic of Collective Action: Public Goods and the Theor=
y
of Groups. <=
st1:City>
Pad=
dock,
Joel. 1998. “Explaining State Variation in Interparty Ideological
Differences,” Political Research Quarterly =
51
(September): 765-780.
Pil=
lari,
Vimala, and Moses Newsome, Jr. 1998. Human Behavior in the Social
Environment:
Families, Groups, Organizations, and Communities.
Presbyterian Layman Committee. 2004. PCUSA Confessing
Churches in
http://www.confessingchurch.ho=
mestead.com/contacts.html
(
Quinnipiac University Polling Institute. 2002.
Abortion. June 5.
http://www.quinnipiac.edu/x660.xml (
Ransford, H. Edward, and Jon Miller. 1983. “Race, =
Sex,
and Feminist Outlooks,”
American Sociological Review <=
/span>48 (1) 46-59.
Reichley, A. James. 1986. “Religion and the Future=
of
American Politics,” Political
Science Quarterly <=
span
style=3D'font-size:11.0pt;font-family:Arial'>101 (1):23-49.
Regnerus, Mark D., David Sikkink, and Christian Smith. 1=
999.
“Voting with the
Christian Right: Contextual and
Individual Patterns of Electoral Influence,” Social Forces 77
(June): 1375-1401.
Roche, Douglas. 1970. “The Challenge to Authority =
in
the Catholic Church.” In
American Mosaic: Social Patter=
ns of
Religion in the
Roh, Jongho, and Donald P. Haider-Markel. 2003. “A=
ll
Politics is Not Local: National
Forces in State Abortion
Initiatives,” Social Science Quarterly 84 (March): 15-31.
Rothenberg, Paula S. 1995. Race, Class, and gender in=
the
Integrated Study, 3rd Ed.
Smidt, Corwin. 1988. “Evangelicals within Contempo=
rary
American Politics:
Differentiating between
Fundamentalist and Non-Fundamentalist Evangelicals,” The Western
Political Quarterly 41 (September): 601-620.
Steinfels, Peter. 2003. A People Adrift: The Crisis of
the Roman Catholic Church in
http://www.quickfacts.census.g=
ov (
Verba, Sidney. 1996. “ The Citizen as Respondent:
Sample Surveys and American
Democracy,” American
Political Science Review 90 (March): 1-7.
Verbit, Mervin F. 1970. “ The Components and
Dimensions of Religious Behavior:
Toward a Reconceptualization of
Religiosity.” In American Mosaic: Social Patterns of Religion in t=
he
Wald, Kenneth D. 2003. Religion and Politics in the <=
/i>
Rowman and Littlefield Publish=
ers,
Inc.
Watt, David Harrington. 2002. Bible-Carrying Christia=
ns:
Conservative Protestants
and Social Power.
Weber, Ronald E., and Willia=
m R.
Shaffer. 1972. “Public Opinion and
American State Policy
Making,”
Willis,
=
and Schuster.
Wood, Richard L. 1999. “Religious Culture and
Political Action,” Sociological Theory
17 (November): 307-332.
Zaller, John. 2001. The Nature and Origins of Mass
Opinion.
University Press.
Zeitlin, Irving M. 2004. The Religious Experience:
Classical Philosophical and Social
Theories.
Appendix (Evangelical Protestant Denominations as Define=
d by
the ARDA, 2000)
Two-Seed-in-the-Spirit Predestinarian Baptists; Allegheny
Wesleyan Methodist Connection; Advent Christian Church; African Methodist
Episcopal Zion Church; American Baptist Association; Amish; Apostolic Chris=
tian
Churches; Apostolic Christian Churches of America, Inc.; Apostolic Lutheran
Churches of America; Associate Reformed Presbyterian Church; Assemblies of =
God;
Baptist General Conference; Baptist Missionary Association of America; Beac=
hy
Amish Mennonite Churches; Berean Fundamental Church; Bible Church of Christ,
Inc.; Black Baptists Estimate; The Brethren Church, Brethren In Christ Chur=
ch;
Barren River Missionary Baptists; Bruderhof Communities, Inc.; Calvary Chap=
el
Fellowship Churches; Central Baptist Association Ministries; Church of God =
in
Christ, Mennonite; Church of God General Conference; Church of God, Anderso=
n,
Indiana; Church of God, Cleveland, Tennessee; Church of God, Mountain Assem=
bly,
Inc.; Church of God of Prophecy; Church of God (Seventh Day); Churches of G=
od,
General Conference; Church of the Lutheran Confession; Church of the Brethr=
en;
Church of the Lutheran Brethren of America; Church of the Nazarene; Christi=
an
and Missionary Alliance; Christian Brethren; Christian Churches and Churche=
s of
Christ; Christ Catholic Church; Christian Reformed Church in North America;
Christian Union; Churches of Christ; Community of Christ; Conservative
Mennonite Conference; Conservative Baptist Association of America; Conserva=
tive
Congregational Christian Conference; Duck River and Kindred Baptists
Associations; Eastern Pennsylvania Mennonite Church; Enterprise Baptists
Association; The Evangelical Congregational Church; The Evangelical Covenant
Church; The Evangelical Free Church of America; Evangelical Lutheran Synod;
Evangelical Mennonite Church; Evangelical Methodist Church; Evangelical
Presbyterian Church; Fellowship of Evangelical Bible Churches; The Fire
Baptized Holiness Church; The Association of Free Lutheran Congregations; F=
ree
Methodist Church of North America; National Association of Free Will Baptis=
ts.
Fundamental Methodist Conference, Inc.; General Six
Principle Baptist; Hutterian Brethren; Independent Fundamental Churches of
America; Independent, Charismatic Churches; Independent, Non-Charismatic
Churches; Independent Free Will Baptist Associations; International Churche=
s of
Christ; International Church of the Foursquare Gospel; International
Pentecostal Church of Christ; Interstate and Foreign Landmark Missionary
Baptists Association; Jasper Baptist and Pleasant Valley Baptist Associatio=
ns;
Landmark Missionary Baptists, Independent Associations and Unaffiliated
Churches; American Association of Lutheran Churches; Lutheran
Church—Missouri Synod; Mennonite Brethren Churches, US Conference; Me=
nnonite
Church USA; Mennonite, other groups; Midwest Congregational Christian
Fellowship; The Missionary Church; National Primitive Baptist Convention, U=
SA;
Netherlands Reformed Congregations; New Hope Baptist Association; New Testa=
ment
Association of Independent Baptist Churches and other Fundamental Baptist
Associations/Fellowships; Old Missionary Baptist Associations; Old Order Am=
ish
Church; Old Order Mennonite; Old Order River Brethren; Old Regular Baptists;
Open Bible Standard Churches, Inc.; Original Free will Baptists; The Orthod=
ox
Presbyterian Church; Pentecostal Church of God; International Pentecostal
Holiness Church; Presbyterian Church in America; Primitive Advent Christian
Church; Primitive Baptists Association; Primitive Baptist Churches, Old Lin=
e;
Primitive Baptists, Eastern District Association of; Primitive Methodist Ch=
urch
in the USA; Progressive Primitive Baptists; The Protestant Conference
(Lutheran); Protestant Reformed Churches in America; Reformed Baptist Churc=
hes;
Reformed Episcopal Church; Reformed Mennonite Church; General Association of
Regular Baptist Churches; Regular Baptists; Seventh-Day Adventist Church; T=
he
Salvation Army; Schwenkfelder Church; Seventh Day Baptist General Conferenc=
e,
USA and Canada; Separate Bapt=
ists
in Christ; Southern Baptist Convention; Southwide Baptist Fellowship; Strict
Baptists; Truevine Baptists Association; Church of the United Brethren in
Christ; United Christian Church; United Baptists; Vineyard USA; Wayne Trail
Missionary Baptist Association; Wisconsin Evangelical Lutheran Synod; The
Wesleyan Church
The Year 2004 Proceedings of the ASSR-SW
The Year 2004
Proceedings of the ASSR-SW