T
AKING
ON
THE
‘G
IANT
S
UCKING
S
OUND
’: O
PINION
R
EFERENTS
, E
LITE
S
TRATEGY
,
AND
THE
D
EBATE
OVER
THE
N
ORTH
A
MERICAN
F
REE
T
RADE
A
GREEMENT
Stacey L. Pelika
University of Wisconsin – Madison
In the final days of the 1993 debate over the North American Free Trade Agreement
(NAFTA), President Clinton challenged Ross Perot, a leading opponent to the agreement, to a
televised debate with Vice President Al Gore. Curiously, criticism of Clinton’s decision came
from both sides of the battle line. Senator Robert Dole (R-KS), a prominent supporter of
NAFTA, stated on Face the Nation that the debate was “a big mistake” because it would put
“Perot back on the front page” (Schieffer 1993). Meanwhile, civil rights leader Jesse Jackson,
who strongly opposed the agreement, told the Washington Post that he was furious about the
debate, noting that “they are resurrecting Perot just when the air was going out of his balloon”
(Edsall 1993).
Why would Jackson object to a high-profile opportunity for someone to voice
disapproval of NAFTA, a position he shared? And why would the Clinton administration choose
a political strategy that both gave voice to the opposition and drew the ire of his policy allies? I
argue that the debate serves as an example of the conscious manipulation of an important and
understudied type of political symbol: the policy opinions of groups and individuals. These
symbols, or opinion referents, include the policy preferences of social groups, organized interest
groups, political parties, and prominent political actors. In this paper, I present a theory of how
political actors – which may include referents themselves, other politicians or groups invested in
a policy, or both – try to strategically shape which opinion referents are attached to a policy and
the effects this process has on public opinion and decision-making. I then test the model using
media and polling data from the NAFTA debate.
Theoretical Foundations
This paper builds on two literatures in American politics: the growing literature about
how political actors consciously attempt to influence mass opinion in the policymaking process,
and a more established line of work on how political symbols shape the ways in which citizens
make sense of and form preferences about politics. I argue that fully understanding the latter is
1
This paper was prepared for presentation at the Annual Meeting of the Midwest Political Science Association,
Chicago, April 2006. The author would like to thank Katherine Cramer Walsh, Joe Soss, and Elizabeth Rigby for
their helpful comments and advice. Inquiries may be sent to ## email not listed ##.
Page 1 of 36 
Convention