. 6
( 11)



Figure 5.1. A schematic representation of the election of 1860 in a two-dimensional policy

and South that might at least remove the threat implied by Dred Scott
and avoid war. It is for this reason that the Republican delegates chose
Lincoln as their presidential candidate. In the election, the Republican
vote increased by 520,000 (over the 1856 ¬gure).1 All ¬fteen states north
of a line from New York to Illinois, gave Lincoln an outright majority of
the electoral college (see Table 5.2).
By clarifying the nature of the southern threat, Lincoln may well have
created a dilemma for members of the northern electorate, since nei-
ther the status quo nor the possibility of war could have been deemed
attractive. As in the earlier “electoral” decision over the rati¬cation of
the Constitution, the degree of risk, associated with the election and the
probable consequences, was very high. The fact that a clear majority of
the northern electorate chose Lincoln (and this choice was re¬‚ected in
a majority of the electoral college) means that a core belief was created
in the electorate over the necessity of a constitutional transformation.
Lincoln was the architect of this transformation. Figure 5.1 reproduces

It is true that the total number of voters increased by about 600,000 between 1856

and 1860, but about 240,000 of these were in the border and southern states.

Lincoln and the Civil War

Figure 2.5, and is intended to indicate that Lincoln™s success in 1860 de-
pended on the transformation of northern electoral beliefs, through the
acknowledgment of the relevance of the second labor (or slave) axis in ad-
dition to the primary land/capital axis that had governed U.S. politics until
Riker (1980) reasoned, on the basis of the chaos theorems, that the
existence of a “social preference cycle” in the electorate meant that the
outcome could be manipulated in an “heresthetical” fashion and it was
the existence of such a cycle that allowed for the success of Lincoln™s
move against Douglas in 1858. Whether there was indeed a “social pref-
erence cycle” in 1860 is impossible to tell. In contrast, Mackie (2001)
has argued that the 1860 election can be interpreted in terms of a sin-
gle dimension, where “latitude is attitude.” That is, the “Upper North”
voted for Lincoln, the “Lower North” for Douglas, the “Upper South”
for Bell, and the “Lower South” for Breckinridge. While there is some
evidence for this view in Table 5.2, a formal explanation can be given in
terms of the electoral model of Chapter 1. An individual, in the North,
who accepts Lincoln™s argument that the expected cost, pT, is very high
would choose Lincoln and the state A1 , at cost qF + rC. Another way of
expressing this is that, for such a voter, Lincoln™s valence, » Lin , would be
much higher than any other candidate™s, and this would induce a vote for
The southern view, in contrast, was that the election destroyed the
“Madisonian” logic of the federal bargain of 1787. A tyrannical major-
ity, the North, had come into being through the institutional device of
the electoral college. This gave the South the constitutional authority to
secede. After the election, an attempt was made in the form of the Crit-
tenden Resolution of January 16, 1861, to allocate all land south of the
Missouri Compromise line of 36—¦ 30 to slave interests. This would have
given the South less than the Dred Scott decision implied, but almost all
that could have been desired by the slave-owning elite. Lincoln™s veto
of the compromise con¬rmed to the South that the North did, indeed,
threaten the institution of slavery. South Carolina was the ¬rst state to
secede on December 29, 1860. As more southern states seceded, the costs
for the remaining states, such as Virginia, of remaining in the Union, obvi-
ously increased. Although institutional rules over the secession vote varied
from state to state, the “belief cascade” so generated in the south led even-
tually to secession by eleven states. Gary™s work on the secession decision
in the southern states (Gary 2004) suggests that the belief in the necessity
of secession was not necessarily supported by a majority in the seceding

Architects of Political Change

states. Nonetheless, I refer to this belief as a “core” or equilibrium belief
in the South, because it was indeed held by those who had the power to
institute the choice.
Lincoln, in his inaugural address of March 4, 1861, asserted that the
fundamental document of the Union was the Declaration of Indepen-
dence. Moreover, the compact implied by the Declaration made secession
unconstitutional. For the Confederacy, the constitutional compact of 1787
was broken because of the tyrannical threat of the North. Because these
beliefs were incompatible, war became inevitable.
Once the issue of slavery had become relevant in the minds of the voters,
then electoral choice in the resulting two-dimensional space became very
different from voting in a single-dimensional policy space. The constitu-
tional transformation that occurred before and after the Civil War was the
consequence of the creation of two incompatible beliefs that depended on
differing interpretations of the constitutional changes of 1776 and 1787.
I argue here that the election of 1860 can only be made intelligible in such
a two-dimensional space. Moreover, once embedded in the polity, these
two dimensions have persisted, although their interpretation may have
varied over time. In the sections that follow, I present an argument that
the slavery dimension became profoundly relevant to the electorate as a
direct consequence of Lincoln™s interpretation of the Dred Scott decision
in the Supreme Court in 1857. In a sense, the Dred Scott decision and
Lincoln™s interpretation of it were necessary causes of the Civil War.2 In
concluding remarks, I also present some hypotheses about the nature of
critical elections, and electoral realignments, that have occurred in U.S.
politics over the very long run.

5.3 dred scott and the supreme court, 1857
Dred Scott was born into slavery about 1800 in Virginia, and in 1830
resided, as property of the Blow family, in St. Louis. After the deaths in
the early 1830s of Elizabeth and Peter Blow, Scott was sold to Dr. John
Emerson. Following his appointment as assistant surgeon in the Army of
the United States, Emerson took Scott to the State of Illinois in 1833, and

An event A is a necessary cause of an outcome B if the absence of A implies the

absence of B. As I interpret Lincoln™s argument, the Dred Scott decision gave an
indication of the South™s intentions. Notice that I do not argue that the Dred Scott
decision, by itself, was a suf¬cient cause.

Lincoln and the Civil War

then to Fort Snelling in Wisconsin Territory. In 1836, Dred Scott married
Harriet Robinson at Fort Snelling, in a civil ceremony. Slave states did
not recognize slave marriages, because slaves were not deemed to have
the right to make contracts. Emerson left the Scotts in Fort Snelling while
he worked in Fort Jesup, Louisiana, where he married Eliza Sanford in
1838. Emerson died in Davenport, Iowa, in 1843, leaving Mrs. Emerson
in St. Louis in full control of his property. The Scotts, who had returned
to St. Louis about 1840, attempted to buy their freedom in 1846. When
Mrs. Emerson refused, the Scotts sued for their freedom. The basis for
the suit was that they had lived in the “free” state of Illinois, and in the
Wisconsin Territory. More precisely, the State of Illinois had been admitted
to the Union in 1818, with a constitution that prohibited slavery. That
constitution was still in effect when Dred Scott lived there from 1833 to
1836. Moreover, Fort Snelling was on the west bank of the Mississippi
River, north of the line 36o 31 . As part of the Missouri Compromise of
1820, slavery was forever prohibited in that part of the Louisiana cession
lying north of the line.
The case of Dred Scott v. Irene Emerson began on April 6, 1846 when
petitions were ¬led in the Missouri Circuit Court in St. Louis. As Fehren-
bacher (1981) notes,

the central question raised in the suit”whether extended residence on free soil
liberated a slave”was not an issue in American politics and had already been
tested many times in the Missouri courts with consistent results (129).

The point here was that because slavery was banned in any territory
covered by the Northwest Ordinance of 1787, then residence by a slave in
such territory implied freedom. Transit by a slave, with the slave™s master,
did not however guarantee freedom.3 However, on the technicality that
Scott™s attorney, Samuel Bay, had not proved that it was Mrs. Emerson
who held Scott in slavery, the jury found for Mrs. Emerson. Bay moved
for a new trial, and Mrs. Emerson relinquished ownership of the Scotts,
leaving them in the custody of the sheriff of St. Louis County.

The Northwest Ordinance of July 1787 applied to the territory of the United States

northwest of the River Ohio. In its sixth article, it speci¬ed: “There shall be neither
slavery nor involuntary servitude in the said territory.” The Constitutional Conven-
tion, later in 1787, stated, “The legislature shall have power to dispose of and make
all needful rules and regulations respecting the territory or other property belonging
to the United States.” It followed that this expansion of congressional power included
the Northwest Territory.

Architects of Political Change

The case was called in January 1850, and eventually the judge ordered
that Dred Scott recover his freedom. Mrs. Emerson™s attorneys then ¬led a
bill of exceptions, setting in motion the appeal procedure to the Missouri
Supreme Court. In the new brief presented to the Court, the question of
the applicability of the Northwest Ordinance and the Missouri Compro-
mise was now raised for the ¬rst time. The background to the case was
the appeal in May 1849 by Thomas Hart Benton, one of the U.S. Sena-
tors for Missouri, to the people of the state to stand by him against the
Missouri legislature. The legislature had passed the “Jackson resolutions,”
supporting Calhoun™s pro-slavery resolutions in the U.S. Senate, and in-
structing Benton to conform to them. Two of the three Missouri Supreme
Court justices were Benton™s political and personal enemies.
The Missouri Supreme Court declared that the judgment of the lower
court be reversed. The circuit court in St. Louis took up the case of Dred
Scott v. John F.A. Sanford in April 1854.4 In his plea of abatement, Sanford
argued that Scott was neither a citizen of Missouri nor even of the United
States, but a slave. The Court held, however, that every person born in
the United States, and capable of holding property, was a citizen having
the right to sue in U.S. courts. If Scott was free, he had the right to sue.
Thus the trial had to go forward to determine if Scott was indeed free.
Scott™s lawyer requested the court to instruct the jury that Scott was free by
virtue of the Ordinance of 1787 and the Missouri Compromise. The judge
refused, and explained to the jury that removal to Illinois only suspended
slavery temporarily. The jury found in favor of Sanford.
That month, President Franklin Pierce, a Democrat, signed into law the
Kansas-Nebraska Act, repealing the slavery prohibition of the Missouri
Montgomery Blair, attorney for Scott, ¬led a brief in December 1854 to
the U.S. Supreme Court, arguing the jurisdictional question that Scott was
indeed a citizen of the United States. For the ¬rst time the constitution-
ality of the Missouri Compromise was considered. Sanford™s attorneys
argued that Congress did not have the authority to proscribe slavery in
the territories, and thus Scott had never been free (Ehrlich, 1979: 129).
The court unanimously ordered the case to be reargued, and considered
the case again in December 1856. In the meantime, the Democrat, James
Buchanan, won the presidential election of November 1856. In his last
State of the Union address in December 1856, the outgoing Democrat
President, Pierce, asserted that Congress did not possess the constitutional

It is still unclear why Mrs. Emerson™s brother, John Sanford, became involved.

Lincoln and the Civil War

power to impose restrictions upon any present or future state of the Union.
Pierce referred here to restrictions on slavery such as those implied by the
Missouri Compromise.
Blair™s brief for Scott, ¬led in December 1856, argued for the consti-
tutionality of the Missouri Compromise, citing thirteen acts of Congress
legislating over slavery in the territories, and fourteen judicial decisions
which recognized this right of Congress. If the argument offered by San-
ford™s attorneys were accepted, then state authority would supersede fed-
eral authority, subjecting Congress to state legislatures.
Ehrlich (1979) suggests that after the arguments, the ¬ve justices from
the slave states decided that the Court could peacefully settle the slavery
issue by declaring the Missouri Compromise unconstitutional. Buchanan,
in his inaugural address, noted that

a difference of opinion had arisen in regard to the point of time when the people of
a Territory shall decide this question [of slavery] for themselves. This is a judicial
question, which legitimately belongs to the Supreme Court of the United States,
before whom it is now pending. (Ehrlich, 1979: 129)

The Opinion of the Court of March 6, 1857 was that people of Scott™s
race were not intended to be granted rights and bene¬ts in the Consti-
tution. Thus Scott was not a citizen of Missouri and not entitled to sue
in federal court. Even so, the Court evaluated the case on its merits, and
concluded that Congress could neither prohibit slavery nor pass any law
depriving a citizen of property. Consequently, despite Scott™s residence in
Illinois and the Wisconsin Territory, he had never been free.5

5.4 the illinois election of 1858
The opinion of the Court, written by Chief Justice Taney, was based on
the conclusion that Dred Scott, as a black, had

no rights under the Constitution and hence no standing to sue in federal court.
The decision about this issue, the most racist and also the longest part of Taney™s
opinion, had virtually no support in either legal precedent or history . . . And the
conclusion was devastating to free blacks, since it deprived them of all federal
rights, including access to federal court. (Drobak, 1993: 239)

Nonetheless, Taylor Blow, the son of Scott™s original master, obtained ownership of

the Scott family and they were emancipated in late 1857. Dred Scott died a year later
in St. Louis.

Architects of Political Change

As Jaffa (2000) comments, “Taney took dead aim at the heart of the
anti-slavery argument when he denied that Negroes were comprehended
in the proposition of human equality in the Declaration” (290).
In a speech as early as June 26, 1857, at the Illinois State House in
Spring¬eld, Lincoln argued that

[T]he Dred Scott decision is erroneous.
If this important decision had been made by the unanimous concurrence of the
judges, and without any apparent partisan bias and had been in no part based
on assumed historical facts which are not really true . . . it then might be, perhaps
would be, factious, nay, even revolutionary, to not acquiesce in it as a precedent.
If [the decision] shall be forced upon the country as a political issue, it
will become a distinct and naked issue between the friends and enemies of the
constitution”the friends and enemies of the supremacy of the laws.
[T]he Dred Scott decision was, in part, based on assumed historical facts which
were not really true . . . Chief Justice Taney, in delivering the opinion of the ma-
jority of the Court, insists at great length that negroes were no part of the people
who made, or for whom was made, the Declaration of Independence, or the Con-
stitution of the United States . . . [But] Judge Curtis, in his dissenting opinion,
shows that in ¬ve of the then thirteen states . . . free negroes were voters . . . and
had the same part in making the Constitution.
(Fehrenbacher, 1989a: 393“5)

For these reasons, Lincoln asserted that the Dred Scott decision was
erroneous and he would do what he could to overrule it.
On June 16, 1858, Lincoln was nominated Republican candidate for
Senator from Illinois, and in his famous acceptance speech predicted that

A house divided against itself cannot stand.
I believe this government cannot endure, permanently, half slave and half
free. . . . I do not expect the Union to be dissolved”I do not expect the house
to fall”but I do expect it will cease to be divided. It will become all one thing, or
all the other. Either the opponents of slavery, will arrest the further spread of it,
and place it where the public mind shall rest in the belief that it is in [the] course
of ultimate extinction; or its advocates will push it forward, till it shall become
alike lawful in all the States, old as well as new”North as well as South.
(Fehrenbacher, 1989a: 426)

Stephen Douglas, the incumbent Democrat candidate for the Senate
position, retaliated on July 9, 1858, by denouncing Lincoln™s “crusade”
against the Supreme Court. In the debates between the two protagonists,
between August 21 and October 15, 1858, Douglas attempted to defend
his “position of popular sovereignty” against Lincoln™s precise attack on
the implications of the Taney Opinion. (The speeches of Douglas and

Lincoln and the Civil War

Lincoln, together with their questions and answers are collected in Holzer,
1993). For Lincoln, this Opinion by the Supreme Court only made sense
if the South intended to extend slavery throughout the Territories, as far
as the Paci¬c. This violated the compromise implicit in the Rati¬cation,
that slavery be accepted, but only in those states that had, by tradition,
been slave states. In Lincoln™s later speeches on the East Coast in 1860,
he seemingly argued that the intention of the South was to extend slavery
throughout the entire Union. I contend that the credibility of Lincoln™s
argument dramatically changed the beliefs, and thus the preferences, of
the Northern electorate.
In the debates at Ottawa and Freeport Illinois, on August 21 and 27,
1858, Lincoln had attacked Douglas for his support of the Taney Opin-
ion, arguing that the Opinion would lead, and was intended to lead, to
the extension of slavery to all states. In referring to the Nebraska bill,
Lincoln said, “[I]f another Dred Scott decision shall come, holding that
they cannot exclude it [slavery] from a State, then we shall discover [why
the particular wording of Territory or State were used in the Nebraska
Bill]” (Fehrenbacher, 1989a: 520).
At the second debate, Lincoln asked Douglas if he would acquiesce in
such a decision by the Supreme Court [asserting that slavery could not
be excluded from a state]. Douglas had earlier declared that the Supreme
Court decisions were binding, and therefore could make no reply compat-
ible with the concept of “popular sovereignty.” Lincoln™s second question
to Douglas, at Freeport, was, “Can the people of a United States Terri-
tory, in any lawful way, against the wishes of any citizen of the United
States, exclude slavery from its limits prior to the formation of a State
Constitution?” (Fehrenbacher, 1989a: 542).
Were Douglas to answer “no” to this, “then he would appear to ca-
pitulate entirely to the southern wing of the party and alienate free-soil
Illinois Democrats” (Riker, 1986: 5).
By answering “yes,” Douglas improved his chances in the Illinois Sen-
ate election, but, by angering southerners, he decreased the likelihood of
winning the 1860 presidential election.
In fact, Douglas answered an emphatic “yes,” even though this logic
appeared counter to the Taney Opinion. His argument, that local regu-
lations would allow the people to make their own choice in this matter,
obviously contradicted federal guarantees of private property (in this case,
of slaves).
At the ¬fth debate at Galesburg on October 7, 1858, Lincoln pursued
the attack. By the Taney Opinion,

Architects of Political Change

“[T]he right of property in a slave is distinctly and expressly af¬rmed in the
Constitution!”. . . Nothing in the Constitution or laws of any State can destroy
a right distinctly and expressly af¬rmed in the Constitution of the United States.
Therefore, nothing in the Constitution or laws of any State can destroy the right
of property in a slave.
(Fehrenbacher, 1989a: 714)

As Fehrenbacher remarks (2001), this “new Court doctrine . . . could
produce a ruling protecting slavery within the northern states as well as
in the western territories” (Fehrenbacher, 2001: 287).
The Chicago Daily Press and Tribune was quick to point out that
the answer that Douglas gave to this question, and his apparent complete
acceptance of the Supreme Court decision on Dred Scott, contradicted one
another. The Tribune argued that the Court™s decision implied that slavery
could not be excluded from the Territories.6 Douglas tried to overcome
this obvious contradiction in his answer by asserting that

slavery cannot exist a day or an hour anywhere, unless it is supported by local
police regulations. . . . Hence, no matter what the decision of the Supreme Court
may be on the abstract question, still the right of the people to make a slave
territory or a free territory is perfect and complete under the Nebraska bill.
(Fehrenbacher, 1989a: 552)

However, the Tribune (August 30, 1858) quoted the Supreme Court™s
assertion that

no tribunal acting under the authority of the United States, whether Legislative,
Executive or Judicial, has a right to deny the bene¬ts of the provisions or guar-
antees which have been provided for the protection of private property against
the encroachments of the government. . . . Since slaves are property, the local leg-
islature has not the right to deny the provisions and is required to furnish the
necessary police regulations to maintain slavery.

In fact, this attempt by Douglas to effect a “compromise” by appealing
to southern interests with his acceptance of Dred Scott, and to use “popu-
lar sovereignty” to appeal to Illinois voters, had already failed. As I noted
above, the editorial of Mobile Register, an in¬‚uential southern journal,
had declared, on August 20, 1858, that Douglas was

Chicago Daily Press and Tribune, August 30, 1858. The point here was that for these

territories, not states, covered by the Northwest Ordinance of 1787, slavery was not
permitted by federal ruling (North and Rutten, 1984). Once the territory became a
state, then its voters could, of course, choose for or against slavery.

Lincoln and the Civil War

in a position to offer the Democrat party the alternative of a probable success in
the next presidential campaign if [the South] accepted the modi¬ed platform he
has prepared for them or of certain defeat and permanent destruction as a party
if they do not. There is ruin to them as a national party in either horn of the
dilemma . . . but there is demoralization as well as disaster in one.

If we read this as a denunciation by the South of Douglas™s attempt at
compromise, then it is clear that he had no alternative but to pursue the
popular sovereignty argument in his attempt to win Illinois.
In the November election, Douglas was returned as a U.S. Senator by
a joint vote of the Illinois House (comprising 40 Democrats and 35 Re-
publicans) and Senate (14 Democrats and 11 Republicans). There was no
popular vote for U.S. Senator; the closest to such a vote was for the State
Treasurer. Using this, the tallies were approximately 125,400 (Lincoln, Re-
publican), 121,600 (Douglas, Democrat), and 5,000 (Buchanan, Demo-
crat).7 The pro-Lincoln Chicago Tribune declared that Lincoln should
have won the race, but was cheated by out-of-date districting. There was
some truth to this allegation of gerrymandering. The Republicans took
49.9 percent of the popular vote for House representatives but their 35
House seats gave them only 46.6 percent of the total, while the Democrats,
with 48.5 percent, took 53.3 percent of the total. In the Senate, thir-
teen members were holdovers from the previous election and split (8,
5) for Democrat, Republican, respectively, while the new members split
(6, 6) for the two parties. Even had the House apportionment been exact,
the Republicans would have been in a slight minority 49 to 51 (see also
Donald, 1995: 228).
What is clear from the election result is that close to half the vot-
ers (principally from southern Illinois) accepted Douglas™ compromise
over popular sovereignty. It seems unlikely, however, that Lincoln™s sec-
ond question was merely a ploy to split the Democrat coalition in the
future presidential election.
Lincoln™s biographer, Donald (1995), observes that Lincoln was bitterly
disappointed, but not surprised by the election result, and quotes him as

I am glad I made the last race. It gave me a hearing on the great and desirable
questions of the age, which I could have had in no other way; and though I now
sink out of view, and shall be forgotten, I believe I have made some marks which
will tell for the cause of liberty long after I am gone. (220)

Holzer (1993: 373). In percentage terms, these are 49.7 percent, 48.2 percent, and

2.1 percent, respectively.

Architects of Political Change

Although Donald and Riker both emphasize the importance of Lin-
coln™s second question at Freeport (because it related directly to the Dred
Scott decision), it was Lincoln™s third question to Douglas that raised the
question of the threat posed by the South to the free states of the North
and West: “If the Supreme Court of the United States shall decree that
STATES cannot exclude slavery from their limits, are you in favor of ac-
quiescing in adopting, and following such decision as a rule of political
In the ¬fth debate at Galesburg, Illinois, on October 7, 1858, Lincoln
had spoken after Douglas, observing that Douglas had essentially ignored
this question, in making the assertion that the Supreme Court could never
make such a decision. Lincoln went on to say, “I did not propound [this
question] without some re¬‚ection, and I wish now to address some re-
marks upon it.” He then quoted from the Constitution that it “shall be
the supreme law of the land, that the Judges of every State shall be bound
by it, any law or Constitution of any State to the contrary notwithstand-
ing” (Fehrenbacher, 1989a: 714). As I have noted above, Lincoln had
declared that the Dred Scott decision rested on the proposition that “the
right of property in a slave is distinctly and expressly af¬rmed in the Con-
stitution!” From this syllogism, Lincoln had deduced that “nothing in the
Constitution or laws of any State can destroy the right of property in a
But, argued Lincoln, this is false.

I believe the entire records of the world, from the date of the Declaration of
Independence up to within three years ago, may be searched in vain for one
single af¬rmation from one single man, that the negro was not included in the
Declaration of Independence.
(Fehrenbacher, 1989a: 702)

The contradiction lay in the falsehood of the premise concerning slav-
ery.9 As Lincoln said, “I believe that the right of property in a slave is not
distinctly and expressly af¬rmed in the Constitution, and Judge Douglas
thinks it is” (Fehrenbacher, 1989a: 714). Lincoln was to return to this

Here I quote the transcript of the speech from the Chicago Daily Press and Tribune,

August 30, 1858. This differs from the version in Fehrenbacher (1989a: 542) in that
STATES is capitalized. I emphasize this point, since Lincoln does not refer here to
the Territories, the subject of the Dred Scott decision but to the constituent States of
the Union.
See Fehrenbacher (2001: 287) for further discussion of the implications of this de-

duction from the Dred Scott decision.

Lincoln and the Civil War

moral contradiction in his speeches in 1860, just before the presidential

5.5 lincoln in new york and new haven
In 1859, there was, of course, no guarantee that Lincoln would be nomi-
nated presidential candidate by the Republican party. The obvious candi-
date was, in fact, the abolitionist William H. Seward, originally of Mas-
sachusetts. The strategic choice for the party would depend on whether
Douglas was nominated by a uni¬ed Democratic Party. However, the
Douglas-Lincoln debates had aroused strong interest in the country, and
Lincoln responded to the resulting controversy by contributing autobio-
graphical information (December 1859) to a journalist (J. W. Fell), who
was interested in writing about Lincoln™s life. Lincoln also prepared a
version of the debates for publication. Moreover, he replied quickly to
an invitation to lecture at the Cooper Institute, New York, in Febru-
ary 1860. Donald suggests that Lincoln™s two-week visit to the East
Coast in early 1860 was necessary if he was to gain support from poten-
tial Republicans who would otherwise vote for Seward (Donald, 1995:
Ch. 9).
Lincoln gave two speeches, “The Address at Copper Institute,” Febru-
ary 27, 1860 (Fehrenbacher, 1989b: 111“30 and Holzer, 2004: 249“284)
and “Speech at New Haven, Connecticut,” March 6, 1869, (Fehren-
bacher, 1989b: 132“50). In these speeches Lincoln addressed two issues:
the ¬rst was the relevance of the Constitution for how slavery was to be
judged; the second was the signi¬cance of the Dred Scott decision for free
The ¬rst aspect of the constitutional issue was the proposition that the
Constitution forbade the federal government from controlling slavery.
Lincoln asserted that Douglas agreed with this proposition (because of
the principle of “popular sovereignty”).
In the Address at Cooper Institite, Lincoln examined the voting of
Congress in 1784, 1787, and 1789 over the Northwest Territory, and in-
ferred that seventeen of the original thirty-nine Founding Fathers (the
signers of the 1787 Constitution) judged this ¬rst proposition false.
Moreover, in Acts of 1798 (Mississippi) and 1804 (Louisiana), Congress
forbade the importation of slaves from foreign ports into the Territories,
and insisted that no slave at all be imported, except by the owner for
his own use as settler. In all, Lincoln judged that at least twenty-three

Architects of Political Change

of the thirty-nine Founders approved of control of slavery in some way
(Fehrenbacher, 1989b: 115).
Lincoln then went on to argue that the Supreme Court based its Dred
Scott decision on the ¬fth amendment (that “no person shall . . . be de-
prived of life, liberty, or property, without due process of law”) while Dou-
glas™ “popular sovereignty” was based on the tenth amendment (“powers
not delegated to the United States by the Constitution . . . are reserved to
the States respectively or to the people”). The two amendments could not
be in logical contradiction to the legal control of slavery by the federal
government because they were debated and approved by Congress (and
rati¬ed by the states) within the same period (1789“92) that Congress did
indeed pass legislation to control slavery.
The second aspect of the constitutional issue was the understanding
of the Founding Fathers as to the essential nature of slavery. Was it to
be seen as a permanent feature of the society, or as a temporary compro-
mise between economic interests? In the speech at New Haven, Lincoln
said, “When men are framing a supreme law and chart of government,
. . . they use language as short and direct and plain as can be found . . .
But the Constitution alludes to Slavery three times without mentioning it

The Migration or Importation of such Persons as any of the States now existing
shall think proper to admit, shall not be prohibited by the Congress prior to the
year 1808, but a Tax or duty may be imposed on such Importation, not exceeding
ten dollars for each Person.
(Article I, Section 9)

As Lincoln observed, “They speak of the “immigration of persons,”
and mean the importation of slaves, but do not say so.”

Representatives and direct Taxes shall be apportioned among the several States
. . . according to their respective Numbers, which shall be determined by adding
to the whole Number of free Persons, including those bound to Service for a Term
of Years, and excluding Indians not taxed, three ¬fths of all other persons.
(Article I, Section 2)

Lincoln pointed out that they say “all other persons,”. . . “when they
mean to say slaves” (Fehrenbacher, 1989b: 142).

Lincoln and the Civil War

No Person held to Service or Labour in one State, under the laws thereof, es-
caping into another, shall, in Consequence of any Law or Regulation therein, be
discharged from such Service or Labour, but shall be delivered up on Claim of the
Party to whom such Service or Labour may be due.
(Article IV, Section 2)

From these ambiguities in the Constitution, Lincoln inferred that

Only one reason is possible, and that is supplied us by one of the framers of the
Constitution . . . they expected and desired that the system would come to an end,
and meant that when it did, the Constitution should not show that there ever had
been a slave in this good free country of ours!
(Fehrenbacher, 1989b: 142)


In forming the Constitution they found the slave trade existing; capital invested
in it; ¬elds depending upon it for labor; and the whole system resting upon the
importation of slave-labor. They therefore did not prohibit the slave trade at once,
but they gave the power to prohibit it after twenty years.

Moreover, “they prohibited the spread of Slavery into the Territories
. . . they considered that the thing was wrong.” (Fehrenbacher, 1989b:
By his interpretation of the meaning of the Constitution as re¬‚ected in
the action and words of the Founders, Lincoln could justi¬ably demand
that slavery be marked, just as the Founding Fathers marked it, as an evil.
“We think that species of labor an injury to free white men-in short We
think Slavery a great moral, social and political evil, tolerable only because
. . . its actual existence makes it necessary to tolerate it” (Fehrenbacher,
1989b: 135).
Its “necessity” lay solely in the “magnitude of [the] subject! One
sixth of our population . . . two thousand million dollars” (Fehrenbacher,
1989b: 134).
The intent of the Dred Scott decision clearly was to extend slavery to
the entire United States. Lincoln does not fully spell out the consequences
of the decision. However, in concluding his speech at New Haven he says:
“I am glad to see that a system of labor prevails in New England under
which laborers CAN strike when they want to. . . . One of the reasons
why I am opposed to Slavery is just here.”10

Fehrenbacher (1989b: 144). Capitals in the original published version.

Architects of Political Change

It would seem obvious that, by this remark, Lincoln implied that, un-
der the Dred Scott decision, slaves could be legitimately imported into
northern free states to work in industrial manufacturing.11 I contend that
Lincoln, in these two speeches, initiated a “belief cascade” in the North.
Slavery was an evil that violated the fundamental beliefs of the Founders
as these were embodied in the Constitution. Moreover, the economic in-
terest of the South, in extending slavery (via the mechanism of the Dred
Scott decision), threatened the way of life of free labor in the North.

5.6 the presidential election of 1860
As I have observed, the Republican choice for presidential candidate in
1860 depended in large degree on the Democrats™ choice. If the Demo-
cratic coalition persisted, and chose Stephen Douglas, then the strategic
choice for the Republicans would be a candidate equally popular in the
West. Lincoln, of Illinois, would be a likely contender. If the Democrats
chose a southerner, such as John Breckinridge of Kentucky (the incumbent
Vice President), then an eastern abolitionist like Seward (Senator, and for-
mer governor of New York) would be a strategic choice. In his speeches
in the East, Lincoln was careful to appear in a somewhat moderate posi-
tion, not abolitionist, but demanding that the status quo, as implicit in the
Constitution, be retained, and that the southern pro-slavery advances be
halted. One of Lincoln™s letters in 1860 suggests that he did not agree with
Seward™s “higher-law” view of slavery. Lincoln felt clearly that Seward™s
intransigence frightened moderate Republicans.
The Democratic National Convention, meeting in April 1860, had been
unable to choose between Douglas and Breckinridge, and adjourned. The
uncertainty over its choice, when the Democracy was to reconvene in sep-
arate conventions in June, meant that neither Lincoln nor Seward were
advantaged. On May 10, the National Union Party (the rump of the
Whigs) nominated John Bell (of Tennessee), with Edward Everett (of Mas-
sachusetts) as the vice-presidential candidate. Because this combination
would do well against Seward, many delegates to the Republican conven-
tion in Chicago in late May hoped for a candidate who (unlike Seward)
would be popular in the West (particularly Illinois, Ohio, and Indiana).
Before moving to nomination, the Republican Party Convention
adopted a platform of seventeen “planks.” The second plank argued

See Jaffa (2000) for further discussion of Lincoln™s arguments on the signi¬cance of

Dred Scott for free labor.

Lincoln and the Civil War

for the maintenance of the Union, as bound by the principles of the
Declaration of Independence and the Federal Constitution. The eighth
denied “the authority of Congress, or a Territorial Legislature”to give
legal existence to slavery in any territory of the United States.” The twelfth
argued for an increase in the tariff “to encourage the development of the
industrial interests of the whole country” and to secure “to the work-
ing man, to mechanics and manufacturers an adequate reward for their
skilled labor and enterprise.” Two further planks demanded “the pas-
sage by Congress of the complete and satisfactory Homestead measure”
and a “railroad to the Paci¬c Ocean” (The New York Times, May 18,
1860). The policy position of the party was quite precisely located at a
compromise position on the capital and land axes, moderately protec-
tionist but expansionist in outlook. On the labor axis, the platform was
obviously anti-slavery. The policy proposals were designed to appeal to
voters of the Northeast who might have voted Whig, as well as to vot-
ers of western states (Iowa, Ohio, and Indiana, for example) who saw
a coincidence between their own “expansionist” interests and those of
Douglas. The choice of presidential nominee was crucial for the Repub-
licans. As Donald (1995: 243) has noted, had a uni¬ed Democratic Party
chosen Breckinridge then the anti-abolitionist Seward would have been
a plausible choice. However, as an easterner, Seward would probably do
poorly in western states, especially if he had to compete against Douglas.
Had Douglas been chosen as the single Democrat candidate, then Lincoln,
with his support in Illinois, would have been the obvious choice.
Because of the uncertainty due to the breakup of the Democratic Con-
vention, Seward gained 173.5 (about 37 percent) of the total delegates,
compared to Lincoln™s 102, in the ¬rst ballot at the Republican Con-
vention. Cameron (of Pennsylvania) took 50.5, while Chase (Governor
of Ohio) took 49, and Bates (a Whig from Missouri) took 48 (Donald,
1995: 250). The other votes were scattered among candidates such as
Fremont (the defeated Republican candidate of 1856). Because Seward
had not obtained the required majority of 233, a second ballot was held
giving Seward 184.5 and Lincoln 181. On the third ballot, Lincoln took
231.5 which increased to 235.5 when some Ohio delegates switched. The
delegates moved to the bandwagon, ¬rst giving Lincoln 364 out of 466,
and then nominating him unanimously. To balance the ticket, Hannibal
Hamlin of Maine (a former Democrat) was selected as vice-presidential
The Democratic Party division in June, between Douglas and Breck-
inridge, meant that the Lincoln-Hamlin combination would do well in

Architects of Political Change

the North. Indeed, the October state election victories of the Republican
Party in Pennsylvania, Ohio, and Indiana suggested that Lincoln had a
good chance of winning the presidential election in November.
As I have noted, Riker (1982: 230) saw Lincoln™s success in the election
in November, 1860, as the culmination of a long progression of strate-
gic attempts by the Whig coalition of commercial interests to defeat the
larger “Jeffersonian-Jacksonian” Democratic coalition of agrarian pop-
ulism. While there is some validity in this perspective, I believe it puts too
much weight on the preferences of the political actors in this drama. In
contrast, I suggest that the crucial feature of the election was the way in
which Lincoln triggered changes in the beliefs of a large portion of the
electorate about the meaning of the U.S. Constitution, and in their beliefs
about the signi¬cance of the Dred Scott decision for free labor.
Lincoln had ¬rst to gain the nomination of his party by defeating Se-
ward. While Seward had political support in the Northeast, he had argued
that there was “an irrepressible con¬‚ict between opposing and enduring
forces,” going on to say that “the United States must and will, sooner or
later, become either entirely a slave-holding nation, or entirely a free-labor
nation” (Foner, 1970: 69). These remarks led some Republican delegates
to view Seward with alarm.
In the run-up to the presidential election, Lincoln™s analysis of the Con-
stitution and of the Dred Scott decision must have been in the minds
of many northern voters. Such swing voters would have cause to reject
the Democrats, and Douglas™s popular sovereignty position, for two rea-
sons. First, they would be persuaded by Lincoln™s arguments that popular
sovereignty entailed an extension of slavery that was constitutionally il-
legitimate. Second, this extension threatened the livelihood of free labor.
However, Lincoln had been careful to distinguish himself from Seward,
and to not imply that con¬‚ict was inevitable. In the New York speech he
had said:

It is exceedingly desirable that all parts of this great Confederacy shall be at peace,
and in harmony, one with another. . . . Even though the southern people will not
so much as listen to us, let us calmly consider their demands, and yield to them, if,
in our deliberate view of our duty, we possibly can. (Fehrenbacher, 1089b: 128)

I interpret Lincoln™s speeches to have been designed to persuade the
northern electorate that the threat of the South (at expected cost, pT) was
real. Under Lincoln this threat would be faced, either by some compromise

Lincoln and the Civil War

measure (at cost qC) or, if the South refused conciliation, possibly through
war (at cost r F ).
In November, 1860, there were four presidential candidates: Lincoln,
Douglas, Breckinridge, and Bell. As Table 5.2 has shown, Douglas took
about a third of the popular vote in twelve of the thirty-three states, all
in the Northeast and West. Bell™s policy of the status quo gave him the
electoral college votes of Virginia, Kentucky, and Tennessee, while Breck-
inridge won all nine states of the deep South. (South Carolina allocated
its electoral college vote by a session of the legislature.) Lincoln won a
majority (over 50 percent) in ¬fteen states, and pluralities in Oregon and
California. Only in the last two states did Douglas and Breckinridge gain
between them a majority. New Jersey was equally divided between Dou-
glas and Lincoln, and split its college vote. With almost 40 percent of
the popular vote, Lincoln won 180 (out of 305) votes in the electoral
Even after the presidential election, but before the counting of the elec-
toral college votes (to take place in February 1861), Lincoln was careful
over possible misinterpretation of his intentions. For a speech drafted for
Lyman Trumbull, given in Spring¬eld, Illinois, on November 20, 1860,
Lincoln wrote:

I have labored in, and for, the Republican organization with entire con¬dence that
whenever it shall be in power, each and all of the States will be left in as complete
control of their own affairs respectively, and at as perfect liberty to choose, and
employ, their own means of protecting property . . . as they have ever been under
any administration.
(Fehrenbacher, 1989b: 186)

Donald suggests that Lincoln may have viewed southern “secession-
ist” responses to his election as a bluff. However, the South Carolina
Legislature unanimously decided on November 10, 1860, to authorize a
state convention to consider future relations between that state and the
Union. Perhaps in response, Lincoln concluded the speech for Lyman
Trumbull on November 20 with the sentence, “I am rather glad of this
military preparation in the South. It will enable the people the more
easily to suppress any uprising there, which their misrepresentation of
purposes may have encouraged” (Fehrenbacher, 1989b: 187).
Between November 1860 and Lincoln™s inaugural speech of March 4,
1861, the Union fragmented, with secession by South Carolina on De-
cember 20, 1860, followed by Florida, Mississippi, Alabama, Georgia,

Architects of Political Change

and Louisiana. Congress had proposed the Crittenden Resolutions, on
January 16, 1861, and Lincoln rejected this attempt at compromise. In a
letter to Seward on February 1, 1861, he wrote, “I am for no compromise
which assists or permits the extension of the institution [of slavery] on
soil owned by the nation” (Fehrenbacher, 1989b: 197).
In his various speeches before the inauguration, Lincoln™s words were
calm and measured, referring to his devotion to the Union, the Constitu-
tion and the liberties of the people.
In his inaugural address on March 4, 1861, Lincoln spelled out his view
of the contract on which the Republic was based:

The Union is much older than the Constitution. It was formed, in fact, by the
Articles of Association in 1774. It was matured and continued by the Declaration
of Independence in 1776. It was further matured and the faith of all the then
thirteen States expressly plighted and engaged that it should be perpetual, by
the articles of Confederation in 1778. And ¬nally, in 1787, one of the declared
objects for ordaining and establishing the Constitution, was “to form a more
perfect union.”. . .
It follows from these views that no State, upon its own mere motion, can
lawfully get out of the Union,”that resolves and ordinances to that effect are
legally void; and that acts of violence, within any State or States, against the
authority of the United States, are insurrectionary or revolutionary, according to
the circumstances . . .
A majority, held in restraint by constitutional checks, and limitations, and
always changing easily, with deliberate changes of popular opinions and sen-
timents, is the only true sovereign of a free people. Whoever rejects it, does,
of necessity, ¬‚y to anarchy or to despotism. Unanimity is impossible; the rule
of a minority, as a permanent arrangement is wholly inadmissable; so that re-
jecting the majority principle, anarchy or despotism in some form, is all that is
left . . .
This country, with its institutions, belongs to the people who inhabit it. When-
ever they shall grow weary of the existing government, they can exercise their
constitutional right of amending it, or their revolutionary right to dismember, or
overthrow it.
I therefore consider that, in view of the Constitution and the laws, the Union
is unbroken.
In your hands, my dissatis¬ed fellow countrymen, and not in mine, is the
momentous issue of civil war . . . You have no oath registered in Heaven to destroy
the government while I shall have the most solemn one to “preserve, protect and
defend” it.12

See Fehrenbacher (1989b: 215“24) for the text. Also see Wills (1999: 184) and

Fletcher (2001: 2) for further discussion.

Lincoln and the Civil War

For the elite in the southern states, secession was legitimate because
the Constitution itself was regarded as the founding compact between
states. Lincoln™s election had destroyed the heterogeneity of the re-
public, creating a tyrannical majority- the North- that directly threat-
ened the South. This violated Madison™s logic in Federalist X. More-
over, Madison had expressed the opinion, in Federalist XLIII that “a
breach committed by either of the parties [to a compact between inde-
pendent sovereigns or states] absolves the other; and authorizes them,
if they please, to pronounce the compact violated and void” (Rakove,
1999: 251).
The share of the popular vote in 1860 for Bell (and his policy of the
status quo) had been signi¬cant in Virginia, Tennessee, Arkansas, and
North Carolina. This suggests that a majority, possibly, of the electorate
in these states, saw no reason to secede. Gary (2004) contends that del-
egate elections and the secession conventions in 1861 were manipulated
by the slave-owning elite to force the secessions that did occur between
February and June 1861. It is, in any case, dif¬cult to understand the
logic of secession. Lincoln had made it clear that their institution would
be preserved, but would not be allowed to envelop other states. It is con-
sistent with the logic of risk discussed here that Lincoln was perceived
as a threat to the expansion of slave-owning elite interests toward the
Paci¬c. To prevent this, secession, and the very considerable risk of war
(implied by Lincoln™s inaugural speech), appeared worth accepting. For
this calculation to be rational it must have been believed that Lincoln
would acquiesce to some compromise that would have given the west-
ern territories to slave interests. For Lincoln, however, the expansion of
slavery would have destroyed the Empire of Liberty crafted by Jefferson.
His argument was accepted by the majority of the northern electorate.
However, it required the terrible cost of Civil War to prove that this be-
lief was a core belief, in the sense that it was held by a winning subset
of the population. It is compatible with my interpretation of the intent
of the Founders, in constructing the constitutional apparatus in 1787,
that Lincoln thus proved to be the decisive risk taker who saved the

5.7 concluding remarks
Prior to 1857, the position of Stephen Douglas, and his principle of pop-
ular sovereignty, would have made perfect sense. However, the issue of

Architects of Political Change

populism, or “state rights,” had become confused because of the inter-
pretation that state rights permitted southern states to maintain their
peculiar institution. It was the nature of the distinctiveness of the “slave”
and “economic” ideologies that led to the incoherence of the Supreme
Court decision over Dred Scott. As I have attempted to show in this
chapter, Lincoln could exploit this incoherence to expose Douglas™ posi-
tion, and make credible the Southern threat to the North. The election of
1860 thus brought the second ideological dimension of civil rights into
The fact that there have been two distinct ideological cleavages in U.S.
politics can be used to clarify the relative positions of the two principal
parties since that time.
Following the Civil War, it was natural for the Republican Party to
move to a pro-capital position on the right of the economic ideological
axis (Bordo and Rockoff, 1996). The Democratic Party, under William
Jennings Bryan, moved simultaneously to a more populist position on the
left of this axis in 1896. However, it was not until 1928 through 1932
that the Democratic Party could include both populist agrarian interests
and urban social liberals.13
The move by L. B. Johnson in the 1960s to create the Great Society in
turn made it possible for the Republican Party to begin to attract social
conservatives, who had, in the past, traditionally chosen the Democratic
Party. The positions of both Democrat and Republican presidential can-
didates by 1992 had become almost identical on the economic ideology
dimension. Populism, in the form of soft money, was no longer a political
issue. As in 1860, the political positions of the two parties in the 1990s
were more clearly differentiated on the civil rights axis.
From this perspective, the elections of 1896, 1932, 1964, and possibly
1980 were all “critical,” because they involved movements by the parties
in response to social and economic quandaries.
In this chapter I have presented the view that the election of 1860 was
the result of a transformation of beliefs in the electorate. This transforma-
tion led eventually to war, and to a transformation in the Constitution,

What I have called the dimension of slavery ideology had of course been transformed

by that time into one involving civil and individual rights. The winning Democratic
coalition of F. D. Roosevelt included southern, socially conservative Democrats as
well as voters both socially and economically liberal. These topics are discussed in
Chapter 6.

Lincoln and the Civil War

the institutional apparatus adopted by the society. If I am correct in my
view of the elections of 1896, 1932, and 1964, they too were preceded by
transformations in electoral beliefs, following on from generally perceived
social quandaries. If voter choice is, in some sense, “stochastic,” then it
is plausible that rapid changes in electoral beliefs are part of a process
during which individual changes in opinion are no longer “independent,”
but are highly correlated. Chapter 6 suggests how models of voting may
be constructed to re¬‚ect this possibility.

Johnson and the Critical Realignment of 1964

6.1 introduction —
It is commonly assumed that politics is inherently one dimensional. In such
a world, theory suggests that political candidates would be drawn into the
electoral center in order to maximize votes. Against such a tendency would
be the motivation of ideological political activists to pull their preferred
candidate away from the center. The balance of electoral incentives and
activist pull creates the “political equilibrium” at any election.
But is politics inherently one dimensional? This chapter continues with
the argument that politics is fundamentally two dimensional. Politics may
appear to be characterized by a single cleavage, but this is because the two
parties themselves “organize” politics along the dimension that separates
them. Party disagreement on one dimension of politics makes that dimen-
sion more salient, while the other dimension is obscured by tacit party
The existence of a submerged or passive dimension of politics trans-
forms the calculus of activist support and electoral preferences. Activists
who are most concerned about the dimension of politics on which par-
ties passively agree constitute a pool of disaffected voters who see no
perceptible difference between the two main parties on the issues that
matter most to them. These disaffected activists often offer a temptation
to vote-maximizing candidates who may see them as the potential margin

Chapter 6 reprints material from Gary Miller and Norman Scho¬eld, “Activists and
Partisan Realignment,” American Political Science Review 97 (2003): 245“60, with
permission of Cambridge University Press. The ¬gures in the chapter are reprinted
from Norman Scho¬eld, Gary Miller and Andrew Martin,“Critical Elections and
Political Realignment in the U.S.A.: 1860“2000,” Political Studies 51 (2003): 217“
40, with permission of Blackwell Publishing.

Johnson and the Critical Realignment of 1964

of victory in a close election. Such a development would be resented and
resisted by party activists who are most concerned about the active di-
mension that differentiates the two parties, and would rather see their
party go down to defeat than re-orient itself to a new set of policy issues
and a new agenda.
Party realignments have long fascinated scholars of American poli-
tics, but no simple model has apparently been suf¬cient to capture the
phenomena. These transformations have been conceived as due to ex-
ogenous change in the economic or international environment. I view
such transformations as the direct consequence of the unrecognized two-
dimensionality of the political space. The premise of this chapter is
that “partisan realignment” cannot be understood without understand-
ing the tension existing between vote-maximizing candidates and policy-
specialized party activists, operating in a two-dimensional strategy space.
Although each election outcome may be the result of an equilibrium in
the context of the contemporary party alignment, over a longer time
span there is a kind of “dynamic stability.” Each temporary political
balance results from candidates™ efforts to re-form the optimal coalition
of party activists. These forces destroy the old equilibrium and create a
new one.
Taking the long view, American politics can be seen as a prolonged,
slow manifestation of the inherent instability of multi-dimensional politics
long posited by rational choice theory. For example, the period in Ameri-
can politics between 1896 and 2000 can be visualized as manifesting two
full re-orientations of party politics along distinct ideological dimensions.

6.2 partisan realignments from 1896 to 2000
Knowing whether a state went for Kennedy or Nixon in 1960 does not
allow one to predict whether that state™s electoral votes went to Bush
or Gore in 2000. Party voting in 1960 was still primarily driven by the
economic cleavage of the New Deal. Income and class variables were
strong predictors of individual voting behavior, with middle-class and
professional homeowners voting Republican, and working class union
members voting Democratic. Democratic candidates since the New Deal
had tried to suppress racial voting in order to appeal simultaneously to
northern liberals and southern segregationists; as a result, the Republican
Party while lukewarm on civil rights, was still arguably more liberal than
the Democratic Party.

Architects of Political Change

By 2000, however, the New Deal party alignment no longer captured
patterns of partisan voting. In the intervening forty years, the Civil Rights
and Voting Rights Acts had triggered an increasingly race-driven distinc-
tion between the parties. Carmines and Stimson (1989) argue that racial
issues had become the dominant cleavage in American politics. Huckfeldt
and Kohfeld (1989) claim that race-based voting had driven out class-
based voting. Racial polarities had come to subsume a variety of other
social issues as well, including abortion, women™s rights, and prayer in
schools. Among whites, church attendance was a primary predictor of
Republican voting (Smith, 2002).
As a result, it is no surprise that a state that went for Kennedy in 1960
was not particularly likely to vote for Gore in 2000. In fact, virtually half
of the states (Nevada and ten southern states) that had voted for Kennedy
voted for Bush in 2000. Similarly, eight New England and upper Midwest
states went from the Republican column in 1960 to the Democrats in
2000. A rede¬nition of party cleavages had happened in the intervening
forty years.
Similarly, it is not possible to predict which states went for Nixon in
1960 based on the state™s choice between Bryan and McKinley in 1896.
Once again, the basis of party voting had changed too much. The Repub-
lican Party™s view of the parties was still largely a relic of the Civil War
trauma. The Democrats were a rural, conservative party, whose candi-
date was a product of rural, Protestant America with no af¬nity for the
growing population of urban immigrants in the cities of the Northeast.
The New Deal coalition of southern, agrarian states with Northern lib-
eral states was not even a remote possibility in 1896. As a result, only
10 of the 22 states that voted for McKinley in 1896 voted for Nixon in
Despite the fact that the partisan status of the states in 1896 does not
predict their partisan status in 1960, and that partisan status in 1960 does
not predict that in 2000, the partisan status of the states in 1896 does pre-
dict the electoral behavior of the states in 2000 remarkably well. If one
were to predict that a Republican state in 1896 would be Democratic
in 2000, and vice-versa, one would get all but 6 states correct. Between
1896 and 1960, eleven Republican upper-Atlantic and upper Midwest
states joined the Democratic coalition, while twelve primarily plains and
border states left the Democratic coalition and joined the Republicans.
Beginning in 1964, California, Oregon and most of the rest of the
Northeast have become reliably Democratic, while the South has become

Johnson and the Critical Realignment of 1964

reliably Republican.1 The composition of these two unpredictable trans-
formations is a nearly perfect reversal of the partisan alignment of 1896.
Urban, cosmopolitan states supported the Republican McKinley in 1896,
and the Democrat Gore in 2000. Virtually all of the rural and southern
states that supported Bryan™s brand of traditional (not to mention funda-
mentalist) values in 1896 supported Bush in 2000.
Furthermore, a state that was more strongly Democratic in 1896 is pre-
dictably more strongly Republican in 2000. Miller and Scho¬eld (2003)
showed that a simple regression of percent Democratic vote in 2000 on
percent Democratic vote in 1896 gives a strongly signi¬cant negative co-
ef¬cient. Indeed, one can get a more accurate prediction with this negative
relationship (r 2 = .37) than by predicting the 2000 election outcome from
the much more recent 1960 election (r 2 = .23).
This fact does not make sense from the perspective of the traditional
literature on realignment. In that literature, new “shocks” to the political
system occasionally come along, and sometimes such a shock results in a
partisan realignment. If this were an accurate model, then each new par-
tisan realignment should result in cumulatively “noisy” transformations
of the partisan alignment. There is no reason to believe that a series of
such realignments would result in the parties effectively “trading places”
(Smith, 2002).
This chapter has several purposes. The ¬rst is to argue that the striking
“mirror image” of the 1896 election in 2000 is understandable through
a two-dimensional spatial model. The two underlying dimensions have
remained remarkably similar for a century and a half, even while the
maneuverings of party leaders (candidates and activists) have resulted in
quite different party positions in that underlying two-dimensional space.
The second purpose is to produce a formal argument that explains
the spatial positions of the parties through time as a result of both
activists and candidates. The literature on the internal organization of
parties (Schlesinger, 1994) emphasizes the differences in goals of party
candidates and party activists”but the traditional spatial modelling liter-
ature regards parties as monolithic rational actors, with a single coherent
goal of winning elections. I argue that this spatial modelling literature

Examining the 1964 election, Burnham (1968: 6) noted that the pendulum had begun

to swing back toward the 1896 cleavage. Indeed he noted both a sharp shift from
the 1960 and preceding elections, and a similarity with such sectionally polarized
elections as the Bryan“McKinley contest of 1896.

Architects of Political Change

cannot account for the peculiar pattern by which parties have transformed
themselves in their search for electoral victory.
Party candidates are motivated primarily to win elections, but use ac-
tivist contributions to increase vote shares. Such candidates will on occa-
sion engage in “¬‚anking” moves so as to enlist coalitions of disaffected
voters, at the risk of alienating some of their traditional activist supporters.
A result of such “¬‚anking” moves, in the early part of this century, was a
shift in emphasis from an underlying social dimension to the economic di-
mension. In the last four decades, elections have seen salience shift from
economics to social policy. The net result of a series of such ¬‚anking
moves, over more than a century, has been to reproduce the partisan
alignment of the end of the nineteenth century, but with the parties in
reversed positions in two-dimensional space.
A third purpose of the chapter is to use the distinction between ac-
tivists and candidates to introduce a concept of “dynamic equilibrium.”
This means that, in a given election, candidates seek to win elections
against a constraint given by the location of the current cadre of party
activists and their preferences; this may or may not involve an attempt
to relocate the center of party gravity by building a coalition with dis-
affected activists. Between elections, current and potential party activists
reconsider the impact of party leaders in the recent election on their own
decisions to join or leave the party activists. The net result of those deci-
sions is, quite possibly, a new con¬guration of party activists constraining
the next party candidates. These “¬‚anking” maneuvers over short-term
horizons give rise, over the long run, to the historically documented par-
tisan realignments of the late nineteenth and twentieth centuries.

6.3 party competition in two dimensions
The model assumes that American voters have preferences that can be
represented in a policy space of as few as two dimensions. The two dimen-
sions underlying partisan competition since the Civil War may be thought
of as an economic and a social dimension (see Figure 6.1). The economic
dimension is one in which the left side represents a pro-redistribution
position articulated by rural, populist, consumerist, and environmental-
ist interests. The right side represents a policy position in opposition to
the use of the state to support redistributional policies from the rich to
the poor. This ideological stance is associated with large businesses and
professionals who oppose redistributional taxes, strict regulation of busi-
ness, and large expenditures on welfare and other social policies in aid of

Johnson and the Critical Realignment of 1964


Liberals Cosmopolitans
Cleavage line

ECONOMIC win set of Rc




Figure 6.1. Illustration of Republican and Democrat economic policy positions in a two-
dimensional policy space.

low-income groups. Throughout the era since the Civil War, this axis has
distinguished pro- and anti-business segments of the population.
As assumed in Chapter 5, the vertical dimension is a social dimension.
The issue most consistently loading on this dimension has been racial,
where “liberalism” has been identi¬ed with policies favorable to blacks
and other minorities. “Social” conservatives oppose the use of govern-
ment, especially the national government, to provide additional bene¬ts
or opportunities to Afro-American minorities, for example. This distinc-
tion between racial liberals and conservatives has had much the same
content since Reconstruction. The debate has involved those who consis-
tently oppose greater political, economic, or legal concessions for racial
minorities, as advocated by the other side (although, of course, the status
quo has changed over this time). This social policy dimension has also
been correlated with other policy issues; racial conservatives from Bryan
through Wallace have been associated with Protestant fundamentalism

Architects of Political Change

and traditional family values. At different times, the issues of women™s
rights, crime, abortion, prayer in schools, the military draft, and anti-
communism have been indicators of this underlying social preference di-
While many policy dimensions are correlated with one ideological di-
mension or the other, the two dimensions are relatively independent. There
are leaders and voters who are both economically and socially conserva-
tive (Region B); these include ¬gures from Grover Cleveland, through the
Taft family, up to and including Robert Dole. There are also politicians
who are both economically and socially liberal (Region D). These include
Bob LaFollette, Hubert Humphrey, George McGovern, and Ralph Nader.
There are also a large number of voters who mix one form of con-
servatism with another form of liberalism. The group in Region A can
be called “populists””social conservatives with a marked antipathy for
big business, who favor policies that bene¬t “the little guy.” These in-
clude notable Southern populists from Huey Long and Wright Patman to
George Wallace. Outside of the South, there are other socially conserva-
tive blue-collar workers who are opposed to big government (Jarvis of the
tax revolt) or hostile both to af¬rmative action and to big business (the
northern blue-collar followers of Patrick Buchanan). Aside from their an-
tipathy to big business, however, this group shares a conservative position
on race, religion, and patriotism issues.
The mirror image of the populists is the group that may be thought
of as “cosmopolitans””a term that is meant to imply a linkage with
national economic interests and an appreciation of social diversity. These
have included Nelson Rockefeller, William Weld of Massachusetts, and
Pete Wilson of California, but also many other urban professional and
business interests who are likely to be in favor of abortion rights, opposed
to prayer in schools, and supportive of some form of af¬rmative action
for minorities. Most recently, these include moderate Republicans who
have identi¬ed with James Jeffords™ exit from the Republican Party.
In a European style proportional representation system, such a two-
dimensional diversity of opinion would no doubt be represented by a
diversity of parties (Laver and Scho¬eld, 1998[1990]). As suggested by
Duverger™s (1954) work, however, the American system of ¬rst-past-the-
post, single-member legislative districts, along with the presidential sys-
tem, is generally prohibitive of a multi-party system. The pressures to
negotiate majority coalitions out of minority positions are strong. While
third parties may play an important role in realignment, they do not do

Johnson and the Critical Realignment of 1964

so by winning a signi¬cant number of seats in Congress, but by serving
as catalysts for change in the main parties™ coalitional structures. (Third
parties are discussed more fully below.)
Because the U.S. system is essentially a two-party system in two dimen-
sions, American political parties have to encompass a broad diversity of
policy opinion in order to win a presidential election. A winning electoral
coalition will have to combine voters who have quite different preferences
on at least one dimension. In Figure 6.1, the traditional position of the
Democratic and Republican Parties during the New Deal alignment is ap-
proximated. The parties are sharply differentiated on the economic policy
dimension, but largely undifferentiated on social policy. As Carmines and
Stimson (1989) note, the Republican Party was, if anything, slightly more
liberal on racial policy up until 1964. Members of the Democratic Party,
in particular, while sharing many elements of an economic ideology, held
sharply different positions on race.
In other words, successful American parties must be coalitions of en-
emies. A party gets to be a majority party by forming fragile ties across
wide and deep differences in one dimension or the other. Maintaining such
diverse majority coalitions is necessarily an enormous struggle against
strong centrifugal forces. Consequently, there will always be an electoral
incentive for the losing party to split the majority party by means of
the suppressed policy dimension, and then woo away some pivotal voters
from the winning party. A partisan realignment, I argue, takes place when
shifting public perceptions of the partisan positions of the two parties lead
some subset of the population to switch voting patterns, creating a new
winning partisan coalition.
When two parties differentiate themselves on the basis of one given
dimension, I call that an active dimension. The very fact that the New Deal
parties differed on economic policy guaranteed that legislative debate,
electoral con¬‚ict, and media attention were focused on that dimension.2
At the same time, the New Deal coalition required strict suppression of

At any time, politics will appear largely one-dimensional because the existing party

activist equilibrium will de¬ne party differences along the dimension that distin-
guishes them. Uni-dimensional models will successfully predict most of the variation
in legislative voting patterns (Koford, 1989). Over time, shifts in the composition of
the party activist coalition will change the policy “meaning” of the partisan cleavage
(Karol, 1999). An attempt to estimate the two dimensions underlying party cleavages
over time is found in Scho¬eld, Miller, and Martin (2003) and Scho¬eld and Sened

Architects of Political Change

racial policy, because Southern Democrats and Northern liberals were
so opposed on that dimension. Franklin Roosevelt™s famous explanation
for why he could not openly support a federal anti-lynching bill (it would
alienate key Southern Democrats in Congress) is a striking example of the
threat that activation of the social policy posed to the fragile Democratic
coalition. Race was a potential fault line within the Democratic Party,
which the Republican opposition tried to exploit. Every party alignment
must create for the minority party a similar opportunity to divide the
majority coalition.
This chapter argues that the reversal of party positions between 1896
and 2000 can only be understood if there are at least two such ideologi-
cal dimensions. At times (e.g., the New Deal), the party cleavage line has
divided economic liberals from economic conservatives. At other times
(e.g., the Reconstruction and the 2000 election), the party cleavage line
has more consistently divided social liberals from social conservatives.
A party candidate, especially a candidate in a party that has lost recent
elections, may try to emphasize a previously suppressed policy dimension.
This may have the effect of both splitting the opposing party™s coalition,
and bringing disaffected voters who have intense preferences about that
suppressed policy into their own party. Success by activating a previously
suppressed dimension of policy may result in pivoting of the party cleav-
age line between the two parties. Two 90—¦ movements (over whatever time
period) of a party cleavage line result in a complete reversal of the ideo-
logical positions of the two parties. This chapter provides both a model
of how this may occur, and an historical interpretation of the two such
movements that have occurred between 1896 and 2000. The model will
take as an illustration the transformation of the New Deal alignment by
means of the activation of the social policy dimension since 1960.

6.4 equilibrium in candidate competition
The literature on party realignment often represents party alignments as
lines of cleavage in a two-dimensional space such as in Figure 6.1. See
Sundquist (1983) for example. This is consistent with the rational choice
literature on spatial modelling, where cleavage lines can be derived from a
“deterministic” model of rational choice by voters. Chapter 2 has empha-
sized that the most important theoretical result from the spatial modeling
literature is that no party may take a position which cannot be beaten by
a challenging party taking some other policy position in the space. Given

Johnson and the Critical Realignment of 1964

any party position such as Da in Figure 6.1, there will in general exist a
set of alternatives, called the “win set,” each of which can defeat Da. The
opposition party, by taking position Rc in the win set of Da should be
able to shift its coalition of support to defeat an incumbent.
Candidates, according to the literature on political parties, are inter-
ested in winning elections (Schlesinger, 1994). An unconstrained chal-
lenging candidate trying to defeat an incumbent identi¬ed at location Da


. 6
( 11)