U.S. Foreign Policy: Long Cycles — What Might They Mean for World Long Cycles?

Jack E. Holmes, Kevin Joldersma and Aaron Keck

International Studies Association

March 19-23, 1998

Note on Attribution

Readers are welcome to cite conclusions of this paper.  However,the exact numbers will not be final until the release of the book, which will explore the variables in this paper along with other variables.

Acknowledgements

Jack E. Holmes would like to thank the Frost Research Center and Hope College for their support of this work.  In addition, he would like to thank several student assistants.  Kevin Joldersma wrote an initial draft of this paper.  Aaron Keck and Brittan Strangways made major editing contributions.  Michael Krukowski and Michael Zolnierowicz worked on the statistical interpretation of the vast amount of data and on editing versions of this work.  Illiana D. Raikova worked on the early statistical foundations, and Dirk Joldersma played a major role in editing work at several points.  In addition, Michael Thelen and Seth Parker are to be thanked for their internet work and research.  Professor Robert E. Elder, Jr. was involved in some joint research that is the basis of portions of this paper.  Professors Tanis, Edward Zajicek, and Roger Nemeth also made suggestions for the statistics in this project.  Sally Smith used her excellent word processing skills to improve the final product.  Responsibility for facts and opinions stated in this paper belong solely to the author.

Introduction

The author has studied U.S. foreign policy long cycles for two decades and has an extensive database covering 1776-1990.  A good deal of attention has been paid to U.S. foreign policy in the post-1945 setting where the U.S. is a world superpower.

Now that the U.S. has been freed from the immediate demands of the Cold War, it is important to study American policy over the long run, especially the period before 1945.  This paper uses conclusions from the author's previous work to raise issues which have implications for the study of world long cycles.

Particular attention is given to consensus (societal/governmental) variables since the U.S. is one of the few countries with a long history under the same written constitution.  The American experience indicates that the U.S. is inclined toward independent action as sometimes dictated by domestic considerations.

While it is clear that the U.S. will act as a superpower in part due to the nature of the international system, it is important to consider some uniquely American features which also have an important impact on policy.

The author's work includes two books on long-term American foreign policy trends.  The first one, The Mood/Interest Theory of American Foreign Policy, was published in 1985.  It presented a general theory regarding how and why introvert and extrovert foreign policy moods alternate every two decades.  This theory, as well as the dates of the cycles, is based on the work of Frank L. Klingberg (1952).  Klingberg elaborates on his cycles in later works (1983 and 1996).

Since the 1985 work contained a good deal of theory, it was thought that a more empirical study of these trends would be in order.  Toward this end, the author created over 150 variables which have been given annual values for at least 150 years.  These variables were based on the work of other authors and compilations of data.  The resulting work, Ambivalent America, which has been the basis for several convention papers and is nearing completion as a book, raises a second set of conclusions which supplement those of the first book.

At the same time, several authors have investigated world long cycles and raised a number of important issues.  Most of these authors emphasize the period since the start of major European influence on the world while others go back prior to that period.  With the exceptions of works by Klingberg and this author, however, similar attention has not been given to American trends.  Pollins and Schweller (1997) do explore some issues relating to interactions of American and world cycles.

The author believes that a conceptual comparison of American long cycles with world long cycles will raise some important issues, and that is the purpose of this paper.

The Mood/Interest Theory

The Mood/Interest Theory of American Foreign Policy argues that there are recognizable patterns in American foreign policy.  According to the author, the U.S. has generally behaved according to six propositions.  These are the public rule proposition, the liberal moods proposition, the interests proposition, the mood/interest conflicts proposition, the executive/legislative proposition, and the extreme mood proposition.  Each is briefly summarized below.

Public Rule Proposition

Public mood is a dominant force in American foreign policy and limits governmental actions.

The mood/interest theory contends that, in the American system of government, the people set the parameters within which the political leadership must act.  The theory does not contend that the American public determines the specifics of individual issues.  Instead, the public guides the decisionmaking process by determining an allowable "channel" for policy makers.  This channel of proper action may have been suggested by the political leadership, but once grasped by the people, it is they who dictate a continuation along this path, regardless of new leadership decisions.

Foreign policy movements within these parameters are acceptable to the public, although short-term public opinion fluctuates on various issues.  Thus, the foreign policy mood parameters are more important to a macroanalysis than is public opinion on specific issues.

Decisions made by foreign policy leaders are expected to be concentrated within the parameters set by the public.  The foreign policy maker is not required to act within the parameters in every instance, but should aim toward results within the parameters, particularly on important issues.   When foreign policy decisions regularly violate the general will of the public, the policymaking officials involved are removed or circumvented through the electoral or political process, and important policies are altered to conform to the public mood.   Such removal and alteration could result from an unrecognized change in the public mood, as depicted in the Liberal Moods Proposition.

Liberal Moods Proposition

American foreign policy moods are expressed and channeled by American liberalism and fluctuate between extreme introversion and extreme extroversion with intervening periods of moderation.

American foreign policy moods are subject to extreme fluctuations best expressed by liberalism and its dislike of concentrated power.  When analyzing problems, Americans are primarily reform liberals (often referred to simply as liberals), business liberals (often referred to in America as conservatives), or some combination thereof.

Although often regarding domestic policy differently, both camps of American liberals are constantly dissatisfied with the harsh realities of international politics and continually try to change them in one of two ways.  One alternative is to minimize active United States participation while setting an example for other nations to follow; this is a basic tendency toward introversion.  The other alternative is to engage in active participation in an effort to change international politics; this is a basic tendency toward extroversion.  Since 1776, American foreign policy has regularly alternated between introvert and extrovert phases, reaching an unrealistic extreme in one orientation before abandoning it for the other.

As noted previously, Klingberg's dates are used until 1940, and 1968-89 is described as the last introvert phase, while 1989 is cited as the start of the current extrovert phase.  This alternation is reflected by a change in the public mood parameters.  The isolationist/internationalist dichotomy is worded to favor the internationalist, just as the globalist/limitationist is worded to favor the limitationist.  Extrovert and introvert are more neutral.  Introvert phases average 21 years while extrovert phases average 27 years as noted in Table 1.

American Foreign Policy: Introvert and Extrovert Years

Table 1

Introvert	Extrovert
1776-98	1798-1824
1824-44	1844-71
1871-91	1891-1919
1919-40	1940-68
1968-89	1989-

Source:  Klingberg (1952) supplemented by Holmes (1985).

Interests Proposition

The United States has generally definable foreign policy interests.

The interests pursued by the United States in the international system must be defined before they can be analyzed.  The discussion of how such interests relate to the mood/interest theory centers on matters of a political and military nature.  The theory acknowledges the secondary importance of economic and humanitarian concerns, but maintains that they are not directly subject to the shifting moods manifested in American liberalism.

The most basic United States interest is the preservation of its territorial integrity and wellbeing.  On a global basis this interest has been reflected in two ways:  attempting to maintain freedom of the seas and, since 1945, preventing a nuclear, chemical, or biological exchange.

The mood/interest theory rejects the argument that American interests are so global that regional variations are of minimal significance.  On the contrary, it asserts that an early point in the identification of interests is the recognition of regional variations.  American foreign policy necessarily treats certain geographical areas as constituting a higher level of interest than others.  These areas include Europe and East Asia, where the United States seeks to prevent one-nation dominance, and the Western Hemisphere, where the United States endeavors to preserve its dominance among major powers.  The United States also seeks to keep other areas of the world, especially the Middle East, open to necessary activity.

The reality/interest zone represents the approximate level of United States activity necessary to guarantee the satisfaction of its fundamental politicomilitary interests.  The time dimension reflects the development of the international system and its relation to United States politicomilitary interests.  As the international community continually increases its ability to challenge United States interests, the level of United States activity necessary to guarantee national interests must rise accordingly.  This is not to say that United States activity will always necessarily increase.

Mood/Interest Conflict Proposition

There is a fundamental conflict between the moods manifested in the liberal American ideology and the dictates of United States politicomilitary interests.

Achieving the interests determined by the harsh realities of the international system usually implies the use of some degree of power.  However, the liberal American orientation to politics includes an aversion to power and, therefore, an aversion to the notion of pursuing politicomilitary interests.  Yet, the extant international situation demands that America pursue its interests to some degree.

Occasionally the United States does too much or too little in pursuit of its interests.  Historically, each mood phase has concluded in unchecked excesses above or below the interest zone.  The problems involved with these excesses result in a shift to the opposite phase.  After the time required to correct previous excesses, each new phase passes through a middle range reality/interest zone in which the liberal mood corresponds to politicomilitary interests.  Eventually, however, the fluctuating mood results once more in excesses beyond realistic foreign policy interests, setting the stage for the next phase.

Executive/Legislative Proposition

Application of the consistent fluctuation in the mood/interest theory can improve understanding of the executive and legislative roles in the determination of American foreign policy.

During extrovert phases the executive plays his strongest role in formulating foreign policy, while the legislature gains power during introvert phases.  During extrovert phases, people are not only willing, but also anxious for the United States to assume an active role in international affairs.  In such times, the sense of urgency implied by stirring slogans allows little patience with the ponderous workings of the legislature.  Thus, foreign policy tends to be delegated to "the man in the best position to know," the chief executive.  The Congress, the historical source of isolationist sentiment and necessarily looking to be reelected, is often content during extrovert phases to let its power slide over to the White House, particularly since the public approves of decisive presidential action.

Introvert phases feature a greater foreign policy role by the Congress.  The congressional response to an introvert public mood is to place greater restrictions on foreign policy activity.  The Congress can best exercise its foreign policy muscle through its ability to say "no."  The president can urge action, but he sometimes needs to accept a more introvert policy than he believes desirable in order to get along with the Congress. Indeed, given the divided structure of American government, a stalemate can lead to an introvert policy given the rising foreign policy interest zone.

Extreme Proposition

The possibility of a return to an extreme form of extroversion or introversion is more likely than is commonly recognized.

Most of the literature that accords some validity to the concept of mood change is optimistic that previous excesses of mood will be avoided in the future.  The mood/interest theory indicates, however, that avoiding extremes will be difficult at best, considering the other propositions.  The result could be that the United States may again fail to do enough or try to do too much to maintain its interests.

Much of the American understanding of the dangers of extremes is based upon experience prior to its superpower status.  Protection of American interests is now more complex simply because more American action is required, a situation resulting in part from strong American action during the 1940-1968 extrovert phase.  Since extremes occur late in a mood phase, it is a very relevant matter now, while time might remain to avoid or mitigate the endofphase extremes of past mood fluctuations.

For the policy maker or scholar seeking relevancy in foreign policy literature, this proposition may have the greatest value.

AMBIVALENT AMERICA

The author has a database with more than 150 variables concerning U.S. foreign policy long cycles.  This database covers a period from 1776 through 1990, and most of the individual variables cover at least 150 years.  Recently, a great deal of attention in the field of international relations has focused on the role of the U.S. in the post-1945 era.  The aim of Ambivalent America is to give needed attention not only to this time period, but also to the period before this key date in history.  This allows us to see U.S. behavioral patterns outside of superpower status.

Prior to 1945, Britain was seen as the world superpower (or hegemon).  Since that time, the United States has been viewed as the premier power in the world.  This makes it especially valuable to study American behavior prior to achieving world power status, because truly =93American=94 behavior can be distinguished from the actions of a hegemon.

Ambivalent America focuses on long cycles in American foreign policy.  Long cycles are the trends that occur with a frequency of at least every decade or two.  More often than not, they are used to study international trends, more than domestic ones.

In particular, five classes of variables are used in Ambivalent America.   Those that are considered most frequently by others are examined most closely.  The five classes of variables are:  American Economic (AE), World Economic (WE), American Consensus (AC), American Military (AM), and World Military (WM).

Ambivalent America concentrates on how these variables relate to American foreign policy mood (FPMOODAM) and American war casualties (CASUALAM).  The mood variable is the one studied in The Mood/Interest Theory of American Foreign Policy, and the casualties variable is reflective of times when the U.S. believes its most vital interests are at stake, as casualties occur during times of war.

The study of the U.S. is important in terms of world long cycles because the U.S. has been a superpower since 1945.  The pressures on a superpower are similar to, but  much more intense than those that other nations face.  These pressures include the need to maintain order in the international system.  This responsibility is extremely important in that the superpower must find a balance between too much and too little action.  If a superpower tries to get involved in too many foreign actions, the result is often war.  However, if a superpower doesn't become involved, and lets events go, the result can be greater international chaos and subsequent war.

Military Variables

The primary finding relating to military variables is that United States foreign policy is influenced by global trends, but does not necessarily act on that basis unless forced to do so.  Generally, the U.S. gets involved in wars only when action is imperative, usually getting involved after war has already begun.  This is especially true prior to 1945.  With respect to hegemonic variables, the United States before 1945 was influenced by British hegemony; after 1945 (the era of American hegemony), United States actions relate to a greater degree with some hegemonic cycles.  Finally, the examination of high/low power concentration variables reveals that high power concentration is prevalent during extroversion, and low power concentration is prevalent during introversion.  Power concentration variables relate more to FPMOODAM than CASUALAM.

In general, CASUALAM relates more to short-term variables and FPMOODAM relates more to long-term variables.  The risk of casualties is far higher during extrovert than introvert periods; indeed the very presence of casualties is indicative of extroversion.  All major wars begin during extroversion, although the Revolutionary War and the Vietnam War extended into the beginning stages of the following introvert period.

Navy variables shed light on important interactions.  During extroversion, world and sea power concentration, as well as the percentage of the U.S. budget spent on the Navy, is larger than during introversion.  During wartime, the number of Navy personnel increases, as does the percentage of global sea power controlled by the United States as other powers lose capabilities.  In general, though, the naval spending variable tells us more about FPMOODAM than CASUALAM; a U.S. naval buildup is indicative of future assertiveness in foreign policy matters, but not necessarily of aggressive military action.

The primary conclusion regarding military variables is that American foreign policy mood and action are most related to domestic considerations, but are also influenced to some degree by world cycles.  Variables relating to naval spending and overall world power concentration relate more to FPMOODAM, as do American variables as a whole; however, variables measuring the degree of international violence tend to relate more with CASUALAM.  A final insight is that variables relating to world leadership cycles, surprisingly, have no significant effect on either macrovariable unless the cycles are modified to fit American trends.

Economic Variables

Nicholas Kondratieff's economic long cycle theories tend to relate strongly to FPMOODAM, but not as much to CASUALAM (perhaps because FPMOODAM is a long-term variable and thus can be expected to relate better to long cycles).  Simon Kuznets' economic long cycles exhibit the same relationship.  In general, long cycles parallel mood cycles; the economy turns up with extroversion and turns down with introversion.

A general finding regarding economic variables is that FPMOODAM relates more with long-term and world economic variables, while CASUALAM relates more with short-term and American variables.  Due to this, short-term world economic variables (specifically, those dealing with inventions and innovations) show little relation to either variable.  Innovation spurts, however, do relate to FPMOODAM, occurring generally in introversion.

In general, prosperity/recovery prevails during extroversion, and recession/depression during introversion.  It is for this reason that inflation is generally higher during extroversion; prosperity and growth trigger inflation.  Additionally, the production levels of the American economy rise during extroversion, as does the money supply.  Also, price levels increase drastically during wartime (probably due to a combination of the increased money supply and the high demands of military interests on the economy).

Some American variables relating to patents, inventions and innovations relate to both macrovariables.  Inventions and innovations generally increase during introversion and peace, and decrease during extroversion and war; however, patents work in the opposite direction, decreasing during peace and increasing during war.

In general, the stock market is stronger during introversion and peace.  The market has long term problems during extroversion, especially during wartime (perhaps because the inherent instability of war and extroversion causes market players to act more warily).  Once a positive outcome of a war is relatively certain, the stock market can head up again.  There are some variables that point toward a stronger stock market during extrovert times, but these generally are short-term indicators.

Consensus Variables

Some variables derived from the first inaugural address of each President relate to both FPMOODAM and CASUALAM.  Presidents serving in extroversion and during casualty periods tend to focus more on power and foreign policy, and less on affiliation.  CASUALAM relates strongly to reelection; a President is more likely to be reelected to office if he serves in a casualty period.  FPMOODAM does not relate to reelection variables unless the Civil War era is excluded.

Other findings are discernable from the institutional and presidential variables.  First, it is more likely that Congress and the presidency are controlled by the same party during extroversion; additionally, more executive agreements are forged during extroversion.  It is more difficult to reach international compromise during times of introversion.  Finally, casualty periods are marked by increases in the number of regulatory acts enacted per year (in response to the necessity of government intervention in a wartime economy).

Societal variables also relate to the two macrovariables.  Great religious awakenings, for example, relate strongly to FPMOODAM.  Additionally, FPMOODAM also relates to generational considerations; activist (civic and idealistic) generations are more likely to be born during extroversion.  Finally, CASUALAM is affected by the ideology of society; a predominantly liberal society is more likely to experience casualties than a conservative one.

Among the major conclusions regarding societal variables is that political consensus is strongest during times of extroversion; consensus may fuel aggressive behavior in extroverted America, leading in some cases to war.  Most institutional variables relate well to CASUALAM, although some (i.e. the President's party in Congress) also relate to FPMOODAM.  Additionally, societal variables relate well to FPMOODAM; this supports the public rule proposition in that the views of society influence foreign policy.  In general, then, extroversion exhibits generational, social, and political cosmopolitanism; a younger political party system; religious revival; annexations; and immigration.  Introversion features parochial generations and conservative ideologies; an older party system; fewer annexations and less immigration; and less religious revival.  Finally, consensus appears to drop off sharply after a war begins.  This evidence indicates that consensus is less important after a war has begun; apparently, consensus is more necessary to begin than to continue a war.

FPMOODAM Over Time

Without going into detail on the findings of each grouping and every time series possibility, there are some clear general trends.  Increases in naval variables generally precede involvement overseas.  At the very least, they signal coming assertiveness in foreign policy.

Economic forces, especially world long cycles, follow foreign policy mood cycles closely.  In general, extroversion signals prosperity, an upturn in the general economic wave.  Introversion, by contrast, signals a downturn.  How much of this decrease is in response to the end of war as opposed to the actual return to introversion is unclear, as introversion and the end of war usually occur rather close to each other.

There is a noticeable difference between American actions in foreign policy before and after 1945.  This is because before 1945, American actions sometimes were necessary in response to the actions of the British hegemon; after 1945, America, as the hegemon itself, was more free to act as it wished.

Several general trends coincide with each other.  Extroversion is generally a time of activism, strong political consensus (which wanes during the middle of a war) and bold new social directions.  Liberalism, economic prosperity, and religious awakening are the norm; party systems are younger; and everyone is willing to take strong steps to solve problems of government and of foreign policy.  These cosmopolitan societies are also more likely to give birth to cosmopolitan, idealistic generations.

CASUALAM Over Time

World military variables relate closer to CASUALAM than American military variables.  American involvement in war is linked with the number of other nations involved in war.  Military variables are more significant in the period after the U.S. became a super power.

American casualties are related more to economic variables which are American in origin than to world variables, but the opposite is true of military variables.  American casualties usually result in a decline in the stock market; the market often improves once it is clear that a war will have a positive result.

A president's reputation and popularity can grow with a successful war.  The president's birth generation is related to the likelihood of war.  Active generations are more war-prone than passive ones.  War is more likely in a time of consensus, but war can also undermine that consensus.

World Long Cycles

The fact that many analysts from wide-ranging times, cultures, and disciplines have identified cycles should make us pause and consider some possibilities.

Long cycles in world history have been studied to a much greater extent than long cycles in American history.  This paper will make use of existing scholarship to examine the ways in which world and American trends interact.  How that interrelation takes place and the relative influence of the American and world spheres remain to be seen.  It is significant that, given its relative isolation prior to 1945 and its great power status since then, the United States has had a good deal to say about its own destiny.  Claiming that America is simply "strong-willed" does not resolve the issue.  Rather, it only indicates how important it is to study American as well as world trends.

A persistent problem in the study of cycles results from the many alternate interpretations.  Crucial questions of dating the beginnings and ends of phases have been vigorously debated.  Certainly, there are legitimate differences of opinion here.  It is beyond the scope of this study to resolve such differences.  Rather it goes with dominant interpretations unless otherwise noted.  Some analysts dismiss cycles by saying that their existence cannot be proven in each and every case.  Others argue that if a cycle has existed in the past, in the absence of proof to the contrary, its continued operation should be presumed.  This study seeks a middle ground between these extremes.

Proponents of long cycles generally agree that since the advent of transoceanic exploration around the beginning of the sixteenth century, there have been five waves of dominant powers, with the present hegemon being the United States.  Each wave has a length of about 100 years, which has sometimes been subdivided into fifty-year and twenty-five-year cycles.  Moreover, for the past several decades, studies have consistently unearthed similarly long-lived economic cycles.  These economic long cycles average about fifty years, a congruence that is hard to overlook.  Some analysts argue that such patterns have been reflected in world realities for thousands of years (Wilkinson, 1992, and Gunder-Frank and Gills, 1992).

Economic Cycles

Among the first modern observers of long-term economic cycles was Nicholas D. Kondratieff.  Kondratieff proposed that economies follow a cyclical or wave-like, rather than linear, pattern.  He devised world business cycles lasting between seven and eleven years each, reflecting price fluctuations, rates of interest, and the production of valuable commodities in the United Kingdom, France and the United States.  However, these cycles do not fully portray the international scene.  To supplement them, Kondratieff shows evidence indicating the existence of smaller, three and one-half-year, and larger, fifty-year cycles (Kondratieff, 1926 [1935]: 105-107).  The fifty-year cycles have attracted the most attention in recent years.  Two of these cycles usually take place within a single hegemonic power cycle.

In Kondratieff's judgment, war and major conflict are symptomatic of economic long waves.  Regular oscillations of the economy create fairly regular societal pressures.  Accordingly, Kondratieff does not see wars as an originator of the long-wave cycles.  He believes that "wars and revolutions, therefore, can also be fitted into the rhythm of the long waves and do not prove to be forces from which these [waves] originate, but rather to be one of their symptoms" (Kondratieff, 1926 [1935]: 112-113). 

Other analysts who have also discovered cyclical recurrences throughout political and economic history support the validity of Kondratieff's theory.  These theorists detect patterns generally similar to those of Kondratieff, although they often differ in their interpretations of the length of the cycles or the reasons for their existence.  Simon Kuznets, for example, argues for a twenty-five to thirty-year up/down economic cycle (Berry, 1991: 65-78).

Leadership Cycles

George Modelski is a pioneer in terms of the long-term patterns of the last five centuries in international politico-military systems (1981a: 63).  He posits one-hundred-year-long phases divided into four internal cycles: global war, world power, de-legitimization, and de-concentration (1981a: 65).  These cycles reflect a "regulatory loop" in which a global war brings about a challenger to the world power.  If successful, this challenger then becomes dominant.  Dominance proves its undoing, however, as the other nations coalesce and arm themselves in response to the newly established concentration of power.  The increased tensions lead to another global war, from which another superpower emerges (Modelski, 1987: 161). 

Characteristics of global powers include military and economic dominance, as well as a globally strategic position.  Modelski defines the latter as having an oceanic location that insures military isolation while promoting oceanic trade.  Modelski has worked closely with William Thompson and Karen Rasler (1994), who link long cycles to naval development, building on the 1948[1962] analysis of Ludwig Dehio.  Another major factor in qualifying for world power status in the Modelski system is a leadership role in the institution of order.  Modelski defines order as "international peace and security, the enjoyment of territorial rights, and the stability of the international economy" (1981a: 64).

Cyclical Interactions

Modelski gives special emphasis to the relationship between economics and politics, stating that "(1) The global economy and the global polity are closely linked through the pivotal role of the world power and through the functions the world power plays in the world economy.  (2) Through these links and mechanisms, the instabilities in the global polity will be transmitted to the global economy and vice versa" (1981a: 71).  Thus, contrary to Kondratieff, Modelski also believes that the structures of the world polity can change the world economy, although he notes that the two do "reinforce each other's instabilities" (Modelski 1981a: 71).

The general analysis in this paper supports two of Modelski's views:  first, that the economic and political cycles move together as one system (Modelski 1981a: 77-79), and, second, that the capacity to innovate is a function of the global system (Modelski, 1981a: 66).  Joshua Goldstein, another researcher of long cycles, agrees, stating that innovations lead price cycles by about five years and that wars are fought after high economic upswings (Goldstein 1988: 256-7, 261).  He proposes three explanations for this war/Kondratieff cycle relationship:

1) The "lateral pressure theory," which states that economic growth within an economic system expands the need of each of its members to gain overseas resources, markets, and spheres of influence;

2) Wars, being costly, take place in times of prosperity when nations believe they can pay the cost; and

3) The "Freudian" perspective that the aggressive mentality promoted by economic prosperity increases the likelihood of expansionist war (Goldstein, 1984: 3-4; Goldstein, 1988: 261-264).

Regardless of the particulars, Goldstein is convinced that there is a relationship between fluctuations in the world economy and the intensity of Great Power wars.  Wars appear to be much more costly and intense during the "upswing" of a Kondratieff cycle.  The ratio of the average annual fatalities of the two periods is 448 to 1!  Of the total 24.7 million deaths during the two periods, 24.6 million happened in the upswings (Goldstein, 1988: 246).

In agreement with Modelski, Goldstein postulates that each economic long wave has witnessed a nation rise to be the dominant figure, both militarily and economically.  Though the two disagree on specific dates, as well as the identity of the first global leader, they agree that world powers and long cycles of world leadership are both in evidence.

Goldstein differs from previously mentioned theorists in his claim that the link between world leadership and world economic long cycles is a tenuous one.  He sees the two cycles as "playing out over time, each according to its own inner dynamic but each conditioned by, and interacting with, the other.  The world system can thus occupy, in rough terms, any position in a two-dimensional space at a given time, depending on the phases of the two cycles" (Goldstein, 1988: 287).  Consequently, each hegemonic wave contains several long waves which climax in war, but it is only after an unspecified number of conflicts that a new hegemonic leader will emerge (Goldstein, 1988: 288).  Brian J. L. Berry (1991) makes an important contribution by placing many long cycles together and analyzing the relationships among them.

This brief introduction can do little more than indicate that there has been a good deal of thought regarding international long cycles.

Case Study:  How Variables Work Together

Now that American and world long cycles have been noted in general terms, it is important to see how they work together in terms of some specific variables.  The variables utilized have been selected on the basis of statistical testing in Ambivalent America.  The purpose of this section is to give a concrete illustration of how trends work together.  This will be done in terms of how trends work in the case of both FPMOODAM and CASUALAM.

General Overview of FPMOODAM

Based on the Ambivalent America analysis, FPMOODAM is an important variable that correlates individually with about half of the Ambivalent America variables.  Logistic regression analysis indicates that 45 variables are especially important.  Fifteen of these are military variables, including variables measuring naval expenditures and the number of American naval personnel, the degree of naval and overall military power concentration among the great powers, and the amount and types of wars started and continuing, especially among great powers.  A higher degree of power concentration, widespread involvement in war, and high intensity of conflict are more characteristic of extroversion.  The seventeen economic variables found in the FPMOODAM regressions include five variations of world long cycles, two American inflation variables measuring trends by decade, innovation spurts, the American stock market, and a measure of world trade barriers.  Upturns in the international economic long cycle, greater American inflation, less innovation, a lower stock market, and lower world trade barriers are characteristic of extrovert phases.  The FPMOODAM regressions also highlight thirteen consensus variables for analysis.  These include variables quantifying national activity, societal unity, and cosmopolitanism from a largely domestic perspective.  For example, the president's political party is usually stronger in Congress during extrovert times than it is during periods of less assertive foreign policy.  During extroversion, society is also more likely to have a generally liberal rather than conservative orientation.

If one works to minimize or eliminate duplication among variables, sixteen variables are especially important.  A graphical time series analysis of these variables suggests that they work together in important ways.

Except for the Revolutionary War and part of the Vietnam War, major war casualties have occurred entirely during periods of extroversion.  In the same way, this section attempts to visually relate extroversion with the many variables shown by this analysis to be significant determinants of public mood.

FPMOODAM Time Variables and Other Variables Mentioned

Table 2

Name	Definition	Years
DGSCSBAC	Degree of Social Cosmopolitanism	1812-1990
FPMOODAM	Klingberg/Holmes Foreign Policy Mood	1776-1990
FPPINGAC	Percentage of First Inaugural Address Devoted to F.P.	1789-1952
GENLDPAC	Assigned Generation of a Person of 35	1776-1990
GENPRZAC	Presidential Assigned Generation	1789-1990
GENYERAC	Birth Generation for the Year (Strauss and Howe)	1776-1990
GLWINTWM	Global War Intensity	1776-1975
GRAWAKAC	Great Religious Awakenings:  Yes/No	1776-1986
HEGMPHWM	World Hegemonic Phase	1776-1984
INFCADAE	Inflation Category by Decade (Batra)	1776-1980
INOVSPWE	Innovation Spurts (Mensch)	1776-1990
INVTUSAE	US Number of World Innovations and Inventions of Cycles)	1776-1938
LCKNVDWE	Kondratieff Cycle (van Duijn)	1782-1973
LCKZABWE	Schlesinger, Jr. Liberal/Conservative Phases and Soete)	1776-1986
NVBUDAAM	Percentage of Budget Spent on the Navy	1794-1990
PRTRECAC	Party Realignment: Up/Down (from Party Realignment C)	1880-1990
PSPTCNAC	Percentage of President's Party in Congress	1789-1976
SEACONWM	Naval Power Concentration	1776-1990
SMCYCLAE	Elliot Stock Market Cycle	1789-1978
TRANUDAM	Transformed Minor Power Dates for the U.S. Up/Down	1776-1988
TRDBLSWE	Trade Barriers of the Leading State (Frederick) 1816-1982 WARTYPWM Type of War	1776-1873

HEGMPHWM is described by four values.  A 1 represents a decline, 2 an ascending power, 3 equals victory, and 4 represents maturity (Wallerstein in Berry, 1991: 161-62).  For the most of U.S. history, American moods and preferences were overshadowed by European powers.  In fact, the United States was a second-rate power throughout the first part of its historical existence.  Changes in FPMOODAM during this time generally follow the hegemonic power phases for the most part, but also reflect American assertiveness more than extra-national affairs.  The rise of the variable in 1815, for example, can be credited to a growth in the power of Great Britain, the leader of the victorious nations after Europe was ravaged by the Napoleonic Wars. 

During the third introvert phase, in the 1870's and 1880's, Great Britain began its decline from the role of hegemon.  At the same time, the American military had been modernized and battle-hardened during the Civil War.  In addition, the United States had earned respect in Europe after its diplomatic successes during the war.

In the subsequent extrovert phase (1891 to 1919), the United States began its ascent toward world hegemony, as many European powers developed colonial empires in Africa and Asia and engaged in an arms race among themselves.  Toward the end of that extrovert phase (1919), the United States made moves toward hegemonic status by helping the Allies to victory in World War I.  The war stretched the European powers' economies and decimated their armies.  The United States can be seen as a direct military beneficiary of this destruction.  Late American entry into the war and geographic isolation meant less damage to U.S. military and economic capabilities and resulted in a relative increase in American power.

The United States was recognized as a first-tier power only near the end of the 19th century after its assertion of power in the Caribbean and Orient.  After World War II, however, the United States matured into its world hegemony.  Starting around the turn of the century, HEGMPHWM starts to reflect American military strength, which becomes preeminent in 1945.  This is not to say that HEGMPHWM is significant only since 1945, only that there are two different forces at work within it for the historical period under consideration.  One reflects the time of Great Britain's dominance; a second, the developments since that time.

As an indicator of power concentration, HEGMPHWM can actually be valuable before 1945.  For that time period it shows how the United States has responded to power concentration, which previous research has indicated as a critical consideration in American decisions to go to war.  When a hegemon is ascending, there is a wider allocation of power in the world.  In other words, there are more nations vying for supremacy, thus making international war more likely.  This threatened one-nation dominance makes war more likely for the United States (Holmes, 1985).

The relationship of the variable to introvert/extrovert phases develops further after World War II, when the United States became the dominant world power.  The level of HEGMPHWM rises during the war, reflecting the growing dominance of the American military around the world.  Power concentration not only reflects relative American power abroad, but it also considers the strength of other nations as well.  The level from 1946 to 1968 represents relative American hegemony around the world.  Levels stay high during the extrovert phase, but begin to decline with the advent of the new introvert phase in 1968.

HEGMPHWM could increase during the current extrovert phase, depending on the degree of American assertiveness overseas.  The United States is likely to fill some of the power vacuum created by the fall of the Soviet Union.  Helping maintain stability in Europe and Eastern Asia and demonstrating leadership in the increasingly independent underdeveloped world would be examples of an expanding American role.  However, nuclear proliferation and the armament of third world powers seriously threaten the existence of further hegemonic phases.

Another measure of power concentration, SEACONWM, (Modelski and Thompson, 1988: 111-12) traces the global concentration of sea power in the number of capital ships.  It shows a rough relationship to FPMOODAM, with several abnormalities.  World concentration of sea power steadily declined, beginning with the end of the War of 1812 and ending in 1940 with only a few spikes stopping it near the end of the British empire.  Although there is some recovery of British naval dominance in the decades after the Civil War, beginning in the 1890's the downward trend reappears.

During this time European parity was achieved through technological advances--dreadnoughts and submarines--and the continuing development of maritime colonies.  A number of European nations competed for naval power during this time, when the importance of sea power was increasingly recognized.  British power rose, albeit slightly, one last time in World War I, but that resurgence was short-lived.  SEACONWM fell during the inter-war period as the world struggled to recover from the economic depression while the navies of the U.S., Germany, and Japan grew.  After a phenomenal rise in American sea power during the Second World War, sea power concentration leveled at about twice its pre-war level.

The variable TRANUDAM (Modelski and Thompson, 1987: 104, 109, 111) measures global leadership cycles, modified in light of American reluctance to assume world leadership after World War I.  TRANUDAM designates world power and de-concentration as 0 (zero), and de-legitimization and global war as 1.  Before 1945, TRANUDAM roughly corresponds with American mood, possibly indicating that the U.S. responded to world instability with its own level of assertion.  (A notable exception occurs during the inter-war years of the 1920's and =9130's, because the source material describes the entire period from World War I to World War II as one of global war.)  After 1945, however, the connection between FPMOODAM and TRANUDAM all but disappears.  This may be because, as a world superpower, the U.S. is more able to act freely, rather than being directed by world trends.  Indeed, geographic isolation from Europe also allowed the U.S. some degree of autonomy even before the turn of the century; however, it is clear that American foreign policy mood coincides more with power phases before 1945.

One war variable proved especially significant to FPMOODAM:  GLWINTWM (Goldstein, 1988: 398), the global war intensity, which measures the intensity of war through the number of fatalities per million Europeans.  Although the wars represented do not always correspond to American involvement, GLWINTWM has only registered war fatalities during introvert periods five times since 1800; and each of these occurred just before or just after an extrovert phase.  This indicates that the foreign policy mood of the United States is closely connected to activity in Europe.

The American Revolution and the War of 1812 were segments of global wars.  While the Revolution was focused in the American colonies, it spread to European nations, ultimately involving France and Spain in addition to England.  Likewise, the War of 1812 was only a minor part of the Napoleonic wars, in which most of the major European nations of the time were involved.  In the mid 1800's, several smaller fluctuations are indicated.  These are, in the order of their occurrence, the Crimean War (1853-56), the War of Italian Unification (1859), the Austro-Prussian War (1866), and the Franco-Prussian War (1870-71).  All of these wars corresponded to the United States' second extrovert mood and bracket the American Civil War.  For the rest of the time, the wars represented are ones in which the United States took an active part, though sometimes belatedly.  Within a half century after the U.S. had settled itself, and resolved its internal conflicts, it turned its attention to defeating totalitarian powers.  This is most pronounced with the involvement in World War II, the Korean War, and the expansion of the Vietnam War.

NVBUDAAM (derived from Historical Statistics and =93Historical Budget Trends=94), the percentage of the total Federal budget devoted to naval expenditures, represents the development of the United States Navy and the priorities of the nation at any particular moment.  Prior to wars involving overseas opponents, NVBUDAAM gradually increases.  This is due to a greater percentage of the budget being diverted to all aspects of the war effort.  Often, it also peaks during the military conflict.  The percent of the budget devoted to the navy then falls off during wartime, most likely due to the requirements of other military services.

The naval variable is quite high at the opening of the 19th century.  As the young nation sought to expand its maritime power, the small budgets allowed any increase to have a disproportionate impact.  Another sharp rise occurs at the beginning of the next extrovert phase.  The Mexican War, a small-scale conflict against a neighbor with little naval power, has a limited effect on NVBUDAAM.  During the Civil War, though, the navy was needed to blockade ports, fight Confederate ships, and project American power to potential British and French threats.  Naval expenditures therefore increased to meet those needs, but the percentage goes down as the nation shifted to a wartime footing and more money was spent on the other armed forces, as it was primarily a land war.

Historically, naval buildups are almost as likely to start during an introvert phase as during an extrovert one.  A naval buildup in an introvert phase, though, is usually followed by extrovert assertion.  However, on the whole, buildups are more likely in extrovert phases.  Generally, there is steady decline in naval expenditures after a major war.  It seems as if the smaller variations are due mostly to the president in office, his view of the necessity of a strong military, and the condition of the American navy--that is, whether it was neglected during an introvert phase.

A final variable that was considered, but not included in time-series analysis, is WARTYPWM (L. L. Farras, Jr., cited in Thompson, 1988: 103-04), the type of wars occurring during a year.  WARTYPWM indicates that hegemonic wars, which mark the changing of world superpowers, and adjusting wars, which are violent but have no significant bearing on the global power status, occur just prior to and mostly during American extrovert phases.  This indicates that the U.S. tends to respond to the more violent wars with its own level of assertion.

Economic Variables

Several versions of the Kondratieff economic long cycle prove relevant to FPMOODAM, and share characteristics including a tendency to rise to economic prosperity after an extrovert phase has begun, possibly indicating that more direct involvement with other nations strengthens the American economy.  Economic long cycles are represented by van Duijn's Kondratieff interpretation, LCKNVDWE (van Duijn, 1981: 268), which generally follows the described pattern, and includes depression (1), recession (2), recovery (3), prosperity (4), and prosperity with war (5).

van Duijn's Kondratieff wave features prosperity in the last decades of the 18th century and then, soon after the beginning of the U.S.'s first extrovert phase (1798), it moves to prosperity with war.  After America's involvement in the War of 1812, it drops to a recession, then becomes a depression one year after the U.S. moves into the next introvert phase.  The economy starts to recover in the last third of that introvert phase (mid-1830's) and moves to prosperity with the onset of the next extrovert phase.  This prosperity remains until the last quarter of this extrovert phase (late 1860's) and then falls to depression at the advent of the third introvert phase.

Again, this previous pattern is repeated with the depression moving to recovery in the final third of the introvert stage and then reaching prosperous levels at the beginning of the next extrovert phase.  The pattern continues, when, after World War I, a time of prosperity with war, it drops to a recession at the end of the extrovert phase and soon moves down to depression.  The pattern breaks somewhat here as the depression lasts to the beginning of the next extrovert phase and then only goes up to a recovery.  Here prosperity is not reached until the end of the Korean War. This last break in the pattern can be explained by the Depression of the 1930's being one of the world's worst in modern times.  Overall, foreign policy mood and the international economy are very much related, with no clear pattern as to which comes first.

Simon Kuznets' inflationary/deflationary economic growth cycle, LCKZABWE (Berry, 1991: 103-05; Berry et al, 1992), shows patterns similar to LCKNVDWE.  It also rises at the beginnings of extrovert phases.  An early rise was seen after the nation achieved independence, as the global and national economy flourished in a climate of free trade.  This relationship shows that both the two and five part interpretations of the world economic long cycles are relevant to this analysis.

The innovation spurts variable (INOVSPWE) (Mensch, 1979: 132) follows its expected pattern.  Generally, such spurts occur in introvert phases coincident to economic downswings.  This is more or less the time when analysts like Schumpeter predict innovation spurts are likely to occur.  The one exception in the American experience is the innovation spurt that started just after the Napoleonic Wars, still during an American extrovert phase, and followed by a comparatively unproductive introvert phase.

The American inflation category per decade variable INFCADAE (Batra, 1987:  188; Holmes and Elder, 1989a: 32) rises early in each extrovert phase and falls before the beginning of the next introvert phase.  The most dramatic instance of this was the falling prices that followed the post-World War I inflation, eventually leading to the Crash of 1929.  Inflation experienced an unusual rise during the late 1830's due to the recovery of the economy after a depression and the Panic of 1837.  Also, decades of high inflation are almost exclusively in extrovert times; the two exceptions are when the United States experienced casualties in introvert times.

The Elliott stock market cycle wave, SMCYCLAE, (Frost and Pretcher, 1981: 20), indicates that the market reaches extraordinary highs or lows during introversion, and corrects itself during extroversion.  The 1837 and 1929 crashes are important instances of this trend.  The variation of the Elliott analysis used to construct this variable pays particular attention to market  weaknesses in the middle third of the 19th century.  Since the majority of stock market variations did not relate to FPMOODAM, it probably is most accurate to state that the relationship between the stock market and FPMOODAM, though weak, indicates downward movements during upward trends in extroversion.

Though the stock market, as indicated by SMCYCLAE, shows constant volatility, there seems to be a slight pattern.  The 4's and 3's of SMCYCLAE, respectively, are the ups and downs (corrections), of the main trend, which in the case of the U.S. is always upward.  The ups and downs in the trend appear nearly continuously throughout American history.  However, more of the downs appear during extrovert phases.  The 1's and 2's of the variable, respectively, represent the downs and ups (corrections) of the corrective wave, which has always been downward.  As with the main trend, the corrective upswings occur most often during extroversion.  Hence, corrections in the general wave tend to occur during extrovert phases.  This is possibly due to the change and uncertainty of the times, which translates to nervous investors and a shaky market.  By contrast, most of the 1's and 4's of the variable occur during introversion, indicating that the market reaches extreme highs and lows during these phases.  This follows this work's previous conclusion that the worst economic times appear in introversion, even though the stock market will remain volatile, no matter what.

Two other economic variables were considered, but not included in the time-series analysis.  INVTUSAE (Freeman, Clark, and Soete, 1982: 178-79), the U.S.'s percentage of world inventions and innovations, indicates an average increase of 6 percent during introversion.  This supports the findings of INOVSPWE; however, the data for INVTUSAE only run through the 1930's, so the impact of new technologies on the variable cannot be assessed.  The second variable, TRDBLSWE (Frederick, 1987: 212-14), measures the trade barriers of the leading state.  These barriers tend to be lower during extroversion and the beginnings of introvert phases, falling further both during and immediately after wars.  This decrease during extroversion and war may be attributable to the fact that, as the nation opens itself to involvement and assertiveness in world affairs, it lowers trade barriers in order to obtain much-needed resources to sustain its aggression.  This is especially true during wartime, which explains the additional decrease.

Consensus Variables

Extrovert moods, which are more often characterized by war than introvert phases, seem more likely to begin during those periods identified by Arthur Schlesinger Jr. as liberal, as measured by the variable LICOJRAC (Schlesinger, Sr., 1949: 81 and Schlesinger, Jr., 1986: 24-5).  This appears to be due to the active nature of liberalism.  A conservative society is unlikely to even consider the sweeping changes inherent in extroversion and war.  The two exceptions to this are the Spanish-American War, which happened immediately following a political realignment and only preceded Schlesinger's liberal phase by a few years, and the Mexican War.  In the first case, President McKinley was pushed, somewhat reluctantly, to declare war by a public pressure that was whipped up by an overreacting media.  President Polk actively promoted the Mexican-American War because of an obligation to the Texans.  Also, Polk had a strong desire to acquire some of the lucrative territory in the Southwest before it fell under the influence of a European power or Mexican power was consolidated or became strong enough to control it.  Significantly, both of these exceptions took place in the beginning thirds of extrovert phases.

The variable FPPINGAC (Klingberg, 1952: 255) measures the foreign policy focus in the first inaugural address of each president.  Presidents who are elected during extrovert phases have more of a focus on foreign policy in their inaugural addresses, while introvert presidents have less. During periods of extroversion, affiliation will be lower.  Likewise, affiliation will be in a stronger position vis-a-vis foreign policy during periods of introversion, when the nation is more intent on resolving domestic crises and reuniting disparate factions.  One notable exception was during the 1920's, which had a high foreign policy content during inaugural addresses while the country was in an introvert phase.  The other major exceptions are during the Civil War and before World War II.  The Civil War may be explained in that the war was about U.S. concerns and therefore, the speech would center on this.  The reason behind the World War II discrepancy is that Franklin Roosevelt gave his first inaugural address in 1933, in the midst of depression and well before any foreign policy crisis.

The Congressional consensus variable, PSPTCNAC (CQ, 1989: 1453), reacts quite strongly to foreign policy mood.  Institutional consensus is high at the beginning of war and lower at the end.  The amount of change is loosely proportionate to the duration of the war.  The Spanish-American War did not affect consensus at all, and other short wars also affected consensus very little.  Longer wars, however, do more damage to consensus.  This is most evident in World War II, when consensus declined after the United States entered the war, and consensus continued to fall for about four years after the war ended.

Another example is Vietnam.  President Johnson had a great deal of partisan support in Congress when troop commitment began, but that support quickly declined, continuing its descent throughout the war and on into the Nixon and Ford administrations.  This was due to a lingering disagreement about whether to continue the war.  It seems that the percentage of the president's party in Congress, though higher in extrovert times, peaks at the end of an extrovert phase.  Following this peak is an almost immediate drop to a very low level at the beginning of the next introvert phase.  This may be due to the changes that an alteration in public mood presents.  The one exception to this general trend of lower party support during introversion is a peak at the end of the introvert phase prior to World War II.

GENYERAC (Strauss and Howe, 1991: 357-65) shows a strong relationship to patterns of foreign policy mood.  GENYERAC classifies years according to the type of Strauss and Howe's four generations being born at that time--reactive, adaptive, civic, and idealist, which are scored 1, 2, 3, and 4 respectively.  While it does not follow FPMOODAM directly, GENYERAC shows a distinct relationship to it.  Shifts in foreign policy mood tend to be followed by similar shifts in assigned generations.  Movements from introversion to extroversion are accompanied by shifts to more aggressive (civic, idealist) generations.  Likewise, moves to introversion bring more passive (reactive, adaptive) generational types.

One of the most difficult aspects of analyzing the Strauss and Howe generations is determining how to assign a year.  Strauss and Howe argue that a person born in a generation carries the values of that generation throughout his or her lifetime.  GENYERAC is the generation of birth for that year, GENLDPAC is the generation of a person thirty-five years old, and GENPRZAC is the generation of the president in office during a particular year.  Birth generations relate most to FPMOODAM, leadership generations most to naval spending, and presidential generations most to casualties.  The interactions of such variables are an important area of future research, although only birth year, GENYERAC, was shown to be statistically significant in relation to FPMOODAM.

The final consensus variable, GRAWAKAC (McLoughlin, 1978: V, 12), measures the incidence of religious revival in American society.  A pattern is difficult to discern, but apparently Great Awakenings coincide with other considerations.  Of the three Great Awakenings recognized in American history, two have happened almost exclusively in extrovert mood phases.  The first awakening followed and the second preceded the shift to extroversion.  The final awakening began in the last third of the 1940 extrovert phase and continued into the following introvert phase.  This anomaly seemed to coincide more with the Vietnam War and continues until the data ends in 1986.

Two other variables were considered, but not included.  DGSCSBAC (Namenwirth, 1973: 661, 674), the degree of social cosmopolitanism in the United States, indicates that extrovert periods are characterized by a high degree of cosmopolitanism.  However, the relationship between the two is weak; DGSCSBAC has a stronger relationship to NVBUDAAM (percentage of naval spending in the federal budget, explained above) than to FPMOODAM.  The second variable, PRTRECAC (Reichley, 1994: A14), examines how close the next political party realignment is to occurring.  Generally, most of the time closest to a new realignment happens during an extrovert phase.

Conclusion

Several overall trends can be seen in the FPMOODAM analysis.  The movements of the military variables show that naval growth often precedes involvement overseas.  A sharp difference can be seen between the period of American history under British hegemony, when the nation's actions were in response to external stimuli, and the post-1945 era, when American actions often catalyzed those of the rest of the world.  Economic forces can often be seen to follow foreign policy mood cycles, usually rising after extroversion begins.  The various theories that have been examined disagree, however, on whether economic declines follow the end of wars or the return to introversion.  Regardless, though, these two events often closely follow one another, so differences are minimal.  Indicators of national consensus tend to rise during extrovert phases, and support for a strong military and possible preparations for war can build as the nation comes together.

Most important is that extrovert public mood tends to occur during times of political consensus and bold new social directions.  Activist presidents governing in prosperous times with a high degree of legislative support tend to be in office during extrovert times, which is also when wars start.  Put in the most basic way, activism and governmental activity often go together in different aspects of national policy, be it  military, economic, or social.  Governmental activity in general appears to open up new directions and possibilities as well as benefit the economy, but has the potential downside of war.  Nobody advocates a war or an unhealthy economy, but both can occur depending on the predominant social orientation in a certain period.

In a general sense, world variables seem quite important in terms of military and economic considerations, but it is clear that American considerations enter the picture in a meaningful way.  While there are no world consensus variables, it is clear that American consensus variables are important to the overall pattern involved.

What is most likely the case is that macro or overall trends work in ways that are not anticipated.  The Democratic Party is most often associated with an active governmental role while Republicans more often favor a more passive governmental role.  The most consistent supporters of a strong laissez-faire economy are in the business community and usually support the Republican Party, while those most vocal in their support of peace and avoidance of international conflict often support the Democratic Party.  Yet, for the last hundred years, Republican presidents have most often been in office when economic times were hardest, and Democratic presidents have most often been in office when major wars started.  Viewed in a continuum, then, governmental activity in the sense of national initiative might be good for the economy and bad for peace.

This is not to formulate a law regarding governmental activity or which party is in power in particular periods, nor is it an argument that people should consider changing party affiliations.  Rather it is to say that the study of long-term trends can provide insight into political/public policy issues.  There could be more than randomness involved in Republican tendencies to get involved in economic difficulty and Democratic tendencies to get involved in international conflict.  While it cannot be proven, a long-term macroscopic perspective presents some evidence that those who are most confident of their abilities to handle certain situations might be the most prone to get into trouble in doing so.  Perhaps Republicans should be more wary of their skepticism against economic intervention at times of crisis, while Democrats should be careful of their support of military intervention at times of crisis even in the hope of bringing about international peace.  Greater caution might improve results in both respects.

CASUALAM Time-Series Analysis

Time-series analysis is an appropriate method to explore how the variables related to CASUALAM work together.  One note of caution needs to be raised:  in some cases CASUALAM itself affects the values of the variables being analyzed.  For instance, wars raise the demand for goods and services, leading to higher inflation.  War can make an unpopular president popular, at least for a while.  The economy, presidential popularity, and many of the other variables analyzed above can also have profound effects on the likelihood of the United States fighting a war in the first place. This section will work to describe how several variables interact together at the same time.

A time-series analysis of variable relationships to CASUALAM offers a different perspective on the conduct of American wars than the FPMOODAM time-series analysis.  FPMOODAM examines introversion and extroversion, which are general indicators of military assertiveness.  In contrast, CASUALAM indicates only those periods when America actually suffered casualties as a manifestation of military involvement.  FPMOODAM examines a broader context, delineating the cycles of public mood in regard to American politico-military involvement.  CASUALAM focuses on actual wars, which take place almost exclusively during extrovert mood phases, giving a specific examination of the domestic causes and effects of U.S. involvement in war.  A comparison of the variables in the CASUALAM regressions in Ambivalent America to those in the FPMOODAM regressions hints at some important differences between the two.

The sequence of CASUALAM logistic regressions left a number of questions to be answered by time series.  These include the following: what are the long-term and short-term effects of casualties on the economy; do American casualties occur after international conflict necessitates it or only serve to incite an already tense situation; how and when do economic changes come into the picture; does the U.S. Navy forecast casualties, reflect casualties, or both.  Time-series analysis should help elaborate on these questions.  At the very least, clear directions for future research will be provided.

Seventy Ambivalent America variables relate directly with CASUALAM, but after consideration of logistic regression and tobit analysis only 36 of these were considered for time-series analysis.  The military variables (nine of the total twelve variables are world variables), mostly relate to the level of violence in the international community.  Two American variables measure the percentage and number of Americans in the navy; the third indicates in which stage of mood the U.S. resides.  Nine of the eleven economic variables considered are American variables.  They describe the stock market, the growth of prices, money supply, and patents; the stock market generally declines while other variables rise.  Overall, American economic variables are more important to CASUALAM than world economic variables describing long cycles.  After logistic and tobit regressions, thirteen American consensus variables were found to be of importance to CASUALAM, including those signifying a liberal political orientation, government control of the economy, executive agreements, and a president who needs little approval from citizens and was born in an actively-oriented generation. 

Nearly equal numbers of military, economic and consensus variables are included in the CASUALAM time-series.  Again, as in the case of FPMOODAM, if several variables reach the same conclusion, only one is used.  A total of thirteen variables are analyzed.  Of these, five relate to military considerations, and four are of an economic nature.  Consensus also shows an important relationship to CASUALAM, with four relevant variables.

CASUALAM Time Variables and Other Variables Mentioned

Table 3

Name	Definition	Years
AGRTRTAC	Ratio of Executive Agreements	1789-1983
BIRTHOAC	Male Birth Order in Family	1789-1986
CASUALAM	Casualty per Year	1776-1990
CASUOLAM	Casualties Yes/No	1776-1990
FPMOODAM	Klingberg/Holmes Foreign Policy Mood	1776-1990
GENPRZAC	Presidential Assigned Generation	1789-1990
GLWINTWM	Global War Intensity	1776-1975
ICVSELAC	Introvert/Conservative-Mixed-Extrovert/Liberal	1776-1986
INFNODAE	Percentage Inflation by Decade (Batra)	776-1979
LCKNMGWE	Kondratieff Cycle (Mager)	1789-1990
LPWCONWM	Long Power Concentration	1776-1990
NVPOPLAM	Percentage of US Population in Navy	1798-1970
POWRDSAC	Power Divergence Score	1789-1990*
PRIINDAE	Wholesale Price Index	1776-1970
PRIINGAE	Percentage Change in Commodity Price Index	1776-1985
PRPACTAC	Passive or Active Orientation	1789-1988
PRPOSTAC	Posterity Ranking	1789-1980
REGACTAC	Number of Regulatory Acts Per Decade	1776-1979
SMPRIMAE	Elliot Stock Market Primary Wave	1791-1976
WARINTWM	International Wars (Annual Nation Months)	1816-1980
WRPICVWM	Percentage of Nations Involved in a Civil War	1816-1980
WRPIINWM	Percentage of Nations Involved in Interstate War	1816-1980

* Indicates some years are missing

Military Variables

Some of the same trends that appear in the FPMOODAM relations to military variables are also apparent in the CASUALAM analysis.  A more distinct relationship can be seen between U.S. casualties and world military indicators during the period of American hegemony.  Before that time, American actions were often subordinate to those of Europe.

The indicator of global war intensity, GLWINTWM (Goldstein, 1988: 398 who cites Small and Singer and Levy), is exemplary in its demonstration of the general pattern.  During the 19th century, America was not involved in most of the great European wars.  (It did, however, participate to a degree in the Napoleonic Wars by fighting the British in the War of 1812.)  Thus, GLWINTWM moves quite independently from American military actions in early U.S. history.  However, with the exception of Vietnam it moves closely with American wars during the 20th century.  The World Wars were bloody and conducted worldwide, while the Korean War was carried on between the camps of the two superpowers, with the United Nations sanctioning the United States side.  The U.S. involvement in Vietnam, though, was not sanctioned by the United Nations, had little effect on worldwide war intensity, and had a comparatively low number of casualties.

American involvement in wars corresponds with the percentage of nations involved in international war, WRPIINWM (Small and Singer, 1982: 118-22).  The first war that can be examined, since the data begins in 1816, is the Mexican-American War.  The wars of the 20th century involved many participants, as did important American wars of the 19th century.  The Mexican-American War and the Civil War both occurred during times when high levels of war existed throughout the world. In both cases, other nations were conducting their own wars while the United States was involved in its hemisphere.  Generally, the percentage of nations involved in war increases immediately before U.S. involvement.  In most cases, except for the Spanish-American War, Korea, and Vietnam, the percentage drops after the U.S. begins to experience casualties.  Often, a higher percentage (or peak) also occurs roughly five years before the peak that signals the beginning of U.S. casualties.

WRPICVWM (Small and Singer, 1982: 252-55), the percentage of nations involved in civil wars, offers interesting insights into American war patterns. It tends to rise during or immediately following most of the American wars, such as the Mexican-American War, Civil War, and World Wars I and II.  No change is seen during the Spanish-American War, which was quite short, and there was a decline in the percentage during the Korean War.  The 1960's and 70's were characterized by a relatively high level of civil war, especially in Third World nations.  This era was also the time of the United States' involvement in Vietnam.  The Cold War made it imperative for both the U.S. and U.S.S.R. to choose sides in such conflicts, in hopes of swinging the balance of power in their favor.  While it is not fair to say that the Cold War incited these civil wars, it may have prolonged them in many cases as a result of great power support for different factions.

Three other variables were considered, but do not appear on the time-series analysis.  WARINTWM (Small and Singer, 1982: 151-54), the incidence of international wars, strongly relates to CASUALAM during the 20th century.  Prior to 1945, the United States typically became involved in international wars two or three years after the outbreak of war (both World Wars are an example).  After 1945, however, a different pattern emerges; the United States, as the leading world power, is more likely to become involved at the start of international conflicts (such as Korea and Vietnam).  WARINTWM's relationship to CASUALAM during the 19th century, however, is unclear.  LPWCONWM (Ray and Singer, 1977; calculated in Modelski and Thompson, 1988: 351-53), which measures long-term power concentration, does not have a unique short-term effect on casualties.  Generally, though, U.S. casualty periods either begin or end in peaks of power concentration.  The War of 1812 and the two World Wars ended in concentration peaks; the Civil War and the Vietnam War began in concentration peaks.  NVPOPLAM (Historical Statistics, 1975: 8, 1141-43; Statistical Abstract, 1989: 335), which measures the percent of the entire national population in the navy, tends to increase at the start of wars.  However, it increases only at the start of those wars that involved overseas powers or required naval assertiveness.  (The Mexican and Spanish-American Wars, for example, did not cause an increase in NVPOPLAM.)

Economic Variables

The economic variables examined in the CASUALAM analysis show a distinct set of patterns that differ greatly from those of the military variables.  Five of the seven economic variables are related to the American economy, whereas six of the seven military indicators were of a global nature.  The economic variables cover a wide range of topics--from long cycles to prices to the stock market--but most exhibit certain similar characteristics.  Most change quite soon after the onset of involvement, indicating that war radically alters the domestic economy.  As with the military variables, a distinct divergence from earlier patterns is seen in the 20th century, especially during the period of American dominance.

Examination of the Kondratieff wave, LCKNMGWE (Mager, 1987: 61) shows quite clearly that wars tend to occur during economic peaks.  Variations can be seen, though, in World War II, which was pressed--possibly against the national inclination--by the continuing aggression of Germany in Europe and the attack on Pearl Harbor, as far as the U.S. is concerned.  American wars occur at all stages of the international economy's expansion: in its beginning (Mexican-American War), at its end (World War I), and in the middle (Civil War).

One American price indice is examined in CASUALAM time series.   PRIINGAE (Historical Statistics, 1975: 199-202; Batra, 1987: 190-91), the percent increase or decrease in commodity prices, rises and peaks during extended wars:  the American Revolution, the War of 1812, the Civil War, World War I, and the Korean War.  The price level then declines after the resolutions of American wars.  (World War II and the Vietnam War are two exceptions; in these two cases, commodity prices did not rise and peak until after the war was over.)  Examining this variable, one can see that the onset of war substantially affects the domestic economy.  Mobilization of the wartime production effort requires regulation of the industry, which, in turn, usually leads to inflation that harms the consumer.  In general, the strongest short-term price increases in American history have immediately followed the onset of war.

The variable INFNODAE (inflation by decade) (Batra, 1987: 188) also relates fairly well with the incidence of war.  Large amounts of money are needed to finance wars.  To meet increasing needs the government often creates more currency.  Once in circulation, this money leads directly to the higher inflation rates seen during war.  Again, the Mexican-American War and the Spanish-American War are exceptions.  Inflation actually declined during the 1840's, mostly due to the banking crisis.  As indicated by other variables, the Spanish-American War was too short to have a large impact on the domestic economy or, therefore, inflation.  In most of the major wars of the U.S., inflation remained at a high level, reinforcing the thesis that war disrupts the domestic economy.  An aberration is the Vietnam War, in which inflation was only at a moderate level.

The effects of war can also be seen in the movements of the stock market, SMPRIMAE (Elliott's Primary Wave) (Frost and Pretcher, 1981).  Examination of the Elliott cycle indicates that most casualty periods in U.S. history are accompanied by corrective down waves in the stock market.  Generally, the stock market is highly variable during wartime, turning downward at the outset of war and recovering as the war's outcome becomes more certain.

Wartime disruption of the world economy usually impacts the U.S. economy.  The United States, because of its close links with the global economy and a tendency to delay involvement in international wars, has often seen its economy affected by war before the nation is participating militarily.  This initial downturn could be attributable to the inherent instability engendered by war, which may make investors more wary and less eager to invest.  Only when the outcome of a war is clearer, and the world begins to regain stability, does the stock market begin to recover.

In most of the major American wars, there has been a turning point which alters the course of the war.  In most cases, the stock market reflects this point, and begins to recover.  In other cases, the final outcome was not certain until after the war was over.  In these wars, the stock market does not recover until after casualties have ceased.  In general, though, the stock market begins to recover as soon as the outcome of the war becomes more certain, whether that be before or after the actual conflict has ended.  In smaller American wars, the forces involved are not quite as clear.

In the 19th century, heated public debate preceded the War of 1812 and the Civil War, each time negatively affecting the stock market.  In the case of the Civil War, the economy was severely disrupted to facilitate wartime production.  The Mexican and Spanish-American Wars did not markedly affect the economy over the long term, but closely followed periods of financial instability--the financial uneasiness of the 1840's and the recession and protectionism of the 1880's and 1890's.  The uncertain back and forth diplomatic maneuvering at the end of the Mexican-American War can be seen in the fall of the stock market in 1848.  Neither war inspired much confidence in the economy, and in both cases the market stayed fairly low.

A huge surge in the stock market is seen immediately following World War II.  The stock market and the economy in general continued to strengthen through the Korean War and the beginning of the Vietnam War.  This rise peaked and began to fall as the introvert phase gained momentum towards the middle of the Vietnam War.  While the stock market fluctuates on a normal basis, the most extreme downturns often accompany a military conflict, though not necessarily one in which the United States is involved.  The uncertainty of war and what lies ahead is reflected in the stock market; this could become even more important in the future due to the global nature of American business.

Consensus Variables

Consensus is strongly related to military conflict, as was shown earlier in this paper.  Eight variables in CASUALAM regressions fall under this category, indicating that the loss of American lives greatly disrupts U.S. society.

Three of these consensus variables relate to the presidency.  The overall conclusion stemming from these relations is that wars enhance a president's reputation.  The likelihood of a president to lead the nation into war can be estimated in several ways.

The variable POWRDSAC (Holmes and Elder, 1989b) is found by subtracting the affiliation score of a president's first inaugural address from its power score.  The breaks in the line are the result of a president who came into office upon the death or resignation of his predecessor, was never re-elected, and thus gave no inaugural address.  In the early 19th century, the power divergence scores were lower, as the nation focused on its expansion and development, rather than on gaining power.  This changed during President Polk's administration, as gaining power became a central objective, and the Mexican-American War followed.  A similar increase in power divergence scores can be seen during the Civil War, as the continuation of the Union was imperative to Lincoln at this time.  The Korean War, following closely on the heels of World War II, shows a decrease in power score during the war, but this is due to the societal reconciliation following World War II.

Affiliation tends to be highest in comparison to power divergence during Schlesinger conservative phases.  Those periods in history include the pre-Civil War era, the Eisenhower era after World War II, and the early days of the Republic, before the turn of the 19th century.

In contrast, those periods with highest power divergence and lowest relative affiliation are the presidencies of Woodrow Wilson and Franklin D. Roosevelt. These were strongly liberal times filled with change.  Wilson led the nation in what was the most violent international war to date and took the first steps toward American hegemony.  His counterpart, Franklin D. Roosevelt, initiated the New Deal, which provided for the most sweeping economic and societal changes the nation had ever seen.  These examples illustrate that, while presidents during wartime are quite power-oriented, presidents immediately after casualty periods are generally more inclined to affiliation.

The variable AGRTRTAC (King and Ragsdale, 1988: 131-40) examines the relationship between the executive and the legislature by measuring the ratio of executive agreements to treaties requiring Senatorial approval.  This variable remained very low until the 20th century, when the nation's international stature rose dramatically.  During the 19th century, the United States sought continued continental expansion rather than international presence.  Therefore the nations of Europe were less interested in the United States, due to its relative isolation and weakness.  Most directly, though, AGRTRTAC remained low during this period because most international agreements went through the Senate.  The legislature was a more powerful force during that century, and the executive branch was seen as having more limited functions than it does today.

When the United States became the hegemonic power, though, AGRTRTAC initially skyrocketed.  During World War II the United States became the dominant player in the Allied camp, and the variable's rise continues during the Korean and Vietnam War eras.  The post-1945 period saw the growth of a more complex international economy, the establishment of the United Nations, and the de-colonizing of Africa and Asia.  All of these events led to dramatic increases in the number of international agreements, which were most often formalized in the United States through executive agreements.  This was also a time of a more assertive executive office, which began to reserve foreign policy to itself.  In regard to conflicts, the pattern that emerges is that the comparative level of executive agreements is high before a war and then drops during the conflict, often recovering after the war's completion.  The exception to this pattern is World War II, in which the level was low before the war, and then rose during the war, reflecting the United States' new role in the international system.

The type of generation the president belongs to can also provide an indication of the nation's likelihood of going to war.  The variable GENPRZAC (Strauss and Howe, 1991: 357-65) classifies presidents as one of four generational types, ranging from the passive and peaceful adaptives to the idealists, who are morally outspoken and often the harbingers of change and sacrifice.  Time-series analysis shows that a president in office during wartime usually comes from a more active generation than that of his predecessor.  Times of war frequently find a government dominated by individuals from idealistic and value-oriented generations.  Woodrow Wilson stands out as an exception to this standard, being a member of an "adaptive" generation known for its willingness to compromise.  These generational characteristics, however, are not commonly given as a description of Wilson.

PRPACTAC (Barber, 1977; Holmes and Elder, 1989b: 541, 555), another presidential variable, indicates whether a president's psychological type was active (1) or passive (0) while in office.  Of the twelve active presidents, five experienced wars.  Once again, the exception was the Spanish-American war occurring during the term of McKinley, regarded as a passive president, as well as the War of 1812.  Since there was no president during the American Revolution, there can be no data regarding this war.  All other U.S. war casualties have occurred during an active presidency.

Four other variables were considered, but not used in the time-series analysis.  REGACTAC (Batra, 1987: 197-213), which measures the number of regulatory acts enacted per decade, indicates that the number of regulatory acts generally increases during wartime.  This is probably due to the necessity of government economic intervention to support a war, particularly in the more complex economy of the 20th century.  BIRTHOAC (Kane, 1981; Holmes and Elder, 1989b: 534, 552), the male birth order of the president, shows that the sitting president during wartime has almost always been a first- or second-born child, supporting the theory that children born earlier are more prone to take decisive action.

The presidential posterity ranking variable, PRPOSTAC (Schlesinger, Sr., 1948 and 1962; Murray and Blessing, 1983), shows that executives who have led the nation to war are generally more highly regarded by historians.  Finally, ICVSELAC (Holmes, 1985; Klingberg, 1952; Schlesinger, Jr., 1986: 24-25; Schlesinger, Sr., 1949: 81), which combines two factors - liberalism/conservatism and introversion/extroversion - indicates that wars begin during extrovert or liberal phases, when the nation is more prone to initiate the sweeping changes involved in war.  So far, each time the U.S. has been both liberal and extrovert, it has suffered war casualties.

While no one variable can be found to accurately show public consensus, it is apparent through the examination of this section that consensus is an important consideration in regard to wars.  When the U.S. becomes involved in a conflict, public consensus is often very high.  As a war continues, and its effects are felt at home, with increased prices, more regulations, and the loss of American lives, consensus declines rapidly.  While wars often begin in periods of economic prosperity, the war can have negative impacts on the economy.  Presidential re-election is a good indicator of how social consensus declines.  President Polk did not run for re-election, because of health concerns and the unpopularity of the Mexican-American War.  Wilson, after World War I, did not realize the change in public mood, and lost his battle with Congress on the League of Nations and the Versailles Treaty.  Probably the best example of how a declining public consensus can damage a president's popularity is President Lyndon Johnson, who, with the growing uncertainty about the Vietnam War, did not run for re-election.  This public dissatisfaction led to Nixon's election.

Conclusion

The analysis conducted in this paper indicates that casualties have significant relationships with long-term military, economic, and consensus indicators. This is less apparent with regard to the military variables, which, before the period of American hegemony, tend to follow European actions.  Economic indicators point to a state of economic expansion at the start of a war, but the economy normally declines as the war progresses.  Again, the stronger and more sustained economy of the post-1945 era is exceptional.  Generally, consensus variables show mixed results.  The level of societal, institutional, and political consensus appears to be a very important consideration prior to a war; consensus usually falls sharply once war begins.  An extrovert/liberal societal outlook characterizes times of war.  In most cases, a president who involves the United States in war is re-elected.  The generational indicator most important to CASUALAM is the president's generation.  This paper shows that in decisions about war and peace, an individual president's priorities and personality can make a difference.

This preliminary time-series analysis indicates that the decision to participate in war and the effects of wartime casualties are complex, multi-faceted issues deserving additional scholarly interest.

Iimplications of American Long Cycles For Study of World Long Cycles

Several cyclical and non-cyclical variables relate to FPMOODAM, the basic U.S. foreign policy mood indicator, and CASUALAM, a measure of American casualties.  It is clear that both are influenced by internal and external considerations.  Variables that prove particularly important include several national and world economic trends, national consensus and active domestic policy, presence or absence of a great power military conflict, and American naval activity.  Innovation spurts, usually occuring in introvert phases, appear to be important to future trends.  Major power military turmoil and domestic considerations are especially relevant to CASUALAM.  Casualties weaken the economy and undermine the consensus.

Many interrelationships may be found between American and world data.  While not all variables are significant to FPMOODAM or CASUALAM, each made a contribution to the study as a whole.  Insight into the long term can only be gained by surveying a broad range of indicators.  This study indicates that variables move in the same general direction at the same time.

American Foreign Policy Moods

The likelihood of American war is closely related to society's current mood and the incidence of war elsewhere in the world.  Society's mood is dependent on a number of considerations.  The economic long waves analyzed by Kondratieff and others show strong relationships to American assertiveness.  World leadership cycles do not significantly relate to FPMOODAM, unless their dates are modified to account for American trends.

Times of extroversion are characterized by a dynamic and violence-prone international climate, stronger economies (due in part to more intense international trade and high wartime demand), and sometimes wavering consensus, which is initially strong when the United States enters a war but may fall as casualties accumulate.  It is during extroversion that the United States affirms its stature in the world community.  Weaker economic indicators, barriers to free trade, and greater worldwide stability, mark introvert times.  These times often stimulate innovation spurts.  The United States seems most concerned with domestic recovery after the costly assertiveness of extroversion.  Thus, both extrovert and introvert periods are necessary: the former to guarantee the nation's international status, and the latter to recover from the strain of such actions.

It needs to be emphasized that the extrovert phase, which this author believes began in 1989, need not be as violent as the 1940-1968 extrovert phase.  Frank L. Klingberg's world consensus cycle, TRICYCWC, indicates a time of less world turmoil in this phase than in most extrovert phases (Klingberg, 1983).  Charles W. Kegley, Jr. and Gregory A. Raymond (1992) maintain that a multi-polar post-Cold War system is not necessarily something to be feared.  Max Singer and Aaron Wildavsky (1993) argue that the strength of democracies in zones of peace makes it quite possible that the United States will have an easier time with foreign policy than in the past. On the other hand, Samuel Huntington (1993) points to a possible clash of civilizations. 

Analysts of public opinion (Richman, 1994 and Holsti, 1994) report that American attitudes favoring world involvement have been quite stable in the past half decade.  In light of the reduced world threat since 1989, the author considers this quite important.  While it is quite clear that the situation is evolving, if long-term considerations are taken into account, then the likelihood of violent situations to develop is smaller.

Relation to Past Research

It is important to relate these conclusions regarding FPMOODAM to concepts in the author's 1985 book.  As Berry (1991) argues, and this author agrees, statistical trends must be buttressed by logical theory. Material from previous work is brought up not to rework old arguments, but to demonstrate that FPMOODAM has the necessary theoretical support.  The author's 1985 book emphasized six propositions as noted earlier in this paper.  First, public mood is important and does set parameters for governmental action.  The analysis of consensus variables demonstrates the importance of a supportive public to foreign policy.  A second proposition emphasizes the importance of alternating foreign policy moods (FPMOODAM), linking them with a Lockean liberalism uncomfortable with the power politics of international relations.  Though the means pursued by extrovert and introvert publics differ, Americans almost always wish the world would operate more like the United States.  Although this work does not present any empirical proof that moods are linked to classical liberalism, evidence of the influence of such liberalism is present.  While classical liberalism is not a perfect expression of American attitudes, it is closer than any other trended constant.  Likewise, classical liberalism is quite comfortable with American ethnic diversity; this could have been developed even more in the author's earlier work.  The third and fourth propositions explain the necessity of meeting minimum politico-military interests, although there can be some mood-inspired deviation in the short term.  This paper shows that through American responses to concentrated global power and major military conflict, these interests have usually been secured.  The fifth proposition, holding that legislative-executive conflict is much more pronounced during introvert phases, is confirmed by this book's institutional consensus study.  The converse, that higher legislative-executive consensus is paired with extroversion, is also demonstrated.  The final 1985 proposition argued that extreme American policy is more likely than it has been commonly recognized.  The evidence in this paper does not indicate a trend away from such extremes.

Fifty-Year Cycles

The strong relationship of world economic long cycles to FPMOODAM is a major finding in this present research, although it was only hinted at in the 1985 book.  What is the reason for these strong relationships? It may be that world economic long cycles generate the momentum behind fifty-year cycles.  It should be noted that both Frank L. Klingberg and the author separately researched American foreign policy moods without knowing how closely they might relate to the Kondratieff cycle.  That closeness became obvious only after the initial work had been done.

One of the reasons why the Kondratieff cycle was not a major part of the initial work was that both Klingberg and Holmes concentrated on American trends to a much greater extent than on world trends.  Berry (1991: 167-75) notes that these fifty-year patterns have appeared throughout history, although he is skeptical of ascribing any particular meaning to this.  Beckman (1983 and 1988) talks about a human desire for cyclic regularity.  The author's conclusions are limited to the observation that world economic and American politico-military cycles might reinforce one another, though they need not be identical in origin or workings.  It could be that, since other countries are more attentive to economic considerations, they respond more assertively to economic threats.  The resulting military challenges to the United States could serve as a stimulant for extrovert foreign policy moods.  But getting away from the theorizing, it is clear that there is a strong statistical relationship between international economic long cycles and American foreign policy moods.

Investigations by two researchers (Kondratieff and Klingberg), who examined two different kinds of data (world economic and U.S.  politico-military), each yielded evidence of a fifty-year cycle.  The Mayans found the same pattern when examining the heavens (Berry, 1991: 167).  Perhaps this was mere chance, but this author believes that more might be involved.  Lembcke (1991-92) fits the fifty-year economic long cycle with the values of the working class.  Critics of the Kondratieff and other cycles raise the valid question:  why fifty years?  Why not some other number?  For the purposes of Ambivalent America other possibilities were considered--for example, Kitchin and Juglar variables--but the author did not uncover any relationships.

This work argues that the underlying mechanism of the 50-year cycle in American foreign policy is a public mood action/reaction continuum.  The cycle begins with Americans, unhappy with the power realities of international relations, trying to change them by minimizing international politico-military involvement.  Minimum activity (or non-intervention) can be rationalized as setting an example, showing other nations a better way.  Unsatisfied with the ultimate results of this passivity, the public charges its leadership with reshaping the world.  Neither orientation works to satisfaction, but it takes two to three generations for this to become clear.  At that point, the first approach is tried again (Holmes, 1985).  Strauss and Howe (1991) also notice generational proclivities toward and away from action. 

What works for a classically liberal country like the United States, with its two-hundred year tradition of democracy, and what might work for a country with another political/social tradition may be quite different.  This study indicates how the public acts when it has a major role in the political system, as is characteristic of the United States and most other democracies.  It may be argued that democracy itself reinforces, or creates, the fifty-year cycles as the public swings from one extreme to another.  Though fifty-year political and public mood cycles are unlikely to be as strong elsewhere as in the classically liberal United States, the fifty-year economic long cycle continually influences all countries.

The dates for international economic long cycles are not the same as those for American foreign policy moods.  Instead they have a "statistically significant relationship," meaning they move in the same directions during generally the same time periods.  This may indicate simultaneous movements of different cycles with different causes.  If one cycle is weaker or wavering, it could be reinforced by a stronger cycle.  If one area of social activity is undergoing major change, it is quite likely that another area will follow.  In other words, multiple cycles might be operating in concert with one another. Identification of which cycle, or cycles, leads the others should be a priority of future research.

American Casualties

Variables related to CASUALAM are generally affected after a rise in casualties, indicating that war causes fluctuations in the variables, rather than the variables causing casualties.  Most economic and consensus indicators fall sharply during wartime.  World trade is disrupted by warfare, and the domestic economy becomes weighed down by the government's wartime needs and regulation.  Although consensus tends to be high prior to war, perhaps reflecting national attempts to justify plans to go to battle, the loss of American lives seems to force the public to rethink that rationale.  It must be acknowledged, however, that World War II may have been a special case where the evil to be opposed was especially clear.

Prices go up rapidly once a war starts, especially if European powers or the United States is involved.  After wars, prices usually go down.  Often a greater concentration of sea power exists when the United States is involved in a war, and the money supply generally expands.  The stock market might improve a bit at the start of a war, but it is likely to go down if the war is prolonged; it often improves once the U.S. looks like a clear winner in a war.  Most central system wars grow to involve the United States.

The analysis in Ambivalent America indicates that consensus and economic variables explain a higher proportion of FPMOODAM than of CASUALAM.  In other words, CASUALAM is most clearly related to military variables, while FPMOODAM is related to military, economic, and consensus variables.  However, evidence shows that both consensus and economic prosperity suffer during times of war, with consensus an earlier victim of war than economic indicators. 

Casualties may change economic and consensus consideration too intricately for explanations based on these two categories of data to fit clear patterns, as is illustrated above.  This is in line with the premise that consensus and prosperity are victims of war.  Analysis of economic variables indicates that casualties have a closer relationship to the American economy than to the world economy.  Even though modern economic understanding can avert some of the wild fluctuations of the past, economic problems are still likely to follow war.  One needs only look at post-World War I Europe and its general recession--which indiscriminately included both victors and the defeated--to detect war costs that linger past the treaties.

Once duplications in the context of variables were eliminated, nineteen variables were subjected to the time series analysis above.  This analysis provides a feasible manner in which variables would work together.  While it is obvious that world military conditions very much affect the American casualties, American as opposed to world economic and consensus variables are most related to casualties.

FUTURE RESEARCH AGENDA

The most important addition that the study of American long cycles makes to the study of world long cycles is the addition of variables concerned with consensus.  This is as can be expected in a democracy.  What is especially interesting about the American democracy is that its structure and values have been stable enough that we have data for a two hundred year period.  Such data will not be easy to find for world long cycles, especially those that include non-Western countries.  What must be noted is that the opinions of society are quite important.  At the very least, there is a need to look at the American experience to see what might be learned about consensus variables.

The American experience gives us some idea about how domestic and international forces might interact.  This interaction suggests that, especially given the increasing interdependence of countries and overall educational levels, governments will have to learn to satisfy their publics.  Since the American government has specialized in making an effort at satisfying the people, much can be learned from its experience.  This is not to deny the obvious problem that one cannot satisfy all of the people all of the time, but rather to say consent of the governed is an important consideration.

A final important consideration is that military, economic, and consensus trends tend to work together.  Thus, it is important to look at the totality of events and trends.  Since it cannot be denied that one overall trend contributes to another overall trend, it is important to look at how they work together.

References

Barber, James D.  1977.  The Presidential Character.  Englewood Cliffs, New Jersey: Prentice-Hall.

Batra, Ravi.  1987.  The Great Depression of 1990.  New York:  Simon and Schuster.

Beckman, Robert.  1983.  The Downwave.  Portsmouth, England:  Milestone Publications.

-------.  1988.  Into the Upwave.  Portsmouth, England:  Milestone Publications.

Berry, Brian J. L.  1991.  Long-Wave Rhythms in Economic Development and Political Behavior.  Baltimore, Maryland and London, England: The Johns Hopkins University Press.

Berry, Brian J. L., Edward Harpham, and Euel Elliott.  1992.  Elections, Party Systems, and the Economic Long Wave.  Paper prepared for the Annual Meeting of the American Political Science Association, Chicago, 5 September.

Congressional Quarterly.  1989.  Guide to the Presidency.  Washington, D. C.: Congressional Quarterly Press.

Dehio, Ludwig.  1948 [1962].  The Precarious Balance.  Trans. Charles Fullman.  New York:  Alfred A. Knopf, Inc.

Frederick, Susanna Y.  1987.  "The Stability of Free Trade."  In Exploring Long Cycles, ed. George Modelski, 186-217.  Boulder, Colorado:  Lynne Rienner Publishers, Inc.

Freeman, Christopher, John Clark, and Luc Soete.  1982.  Unemployment and Technical Innovation.  London, England:  Francis Pinter.

Frost, A.J. and R.R. Prechter, Jr.  1981.  Elliot Wave Principle:  Key to Stock Market Profits.  Second edition.  New York:  Haddon Craftsman.

------.  1996.  Elliott Wave Principle:  Key to Market Behavior.  Eighth edition.  Gainesville, Georgia:  New Classics Library.

Goldstein, J.S.  1984.  "War and the Kondratieff Upswing."  Paper presented at the Annual Meeting of the International Studies Association, Atlanta, 27-31 March.

------.  1988.  Long Cycles: Prosperity and War in the Modern Age.  New Haven, Connecticut:  Yale University Press.

Gunder-Frank, Andre, and Barry K. Gills.  1992.  "The Five-Thousand Year World System:  An Interdisciplinary Introduction."   Paper prepared for delivery at the Annual Convention of the International Studies Association, Atlanta, 2 April.

Historical Statistics of the United States:  Colonial Times to 1970.  1975.  Bicentennial ed. 2 pts.  House doc. 93-78.  U.S. Department of Commerce, Bureau of the Census.  Washington, D.C.: Government Printing Office.

Holmes, Jack E.  1985.  The Mood/Interest Theory of American Foreign Policy. Lexington, Kentucky:  University Press of Kentucky.

Holmes, Jack E., and Robert E. Elder, Jr.  1989a.  "Prosperity, Consensus and U.S. Foreign Policy Assertiveness."  Paper delivered at the Annual Meeting of the International Studies Association, London, 27 March-1 April.

------.  1989b.  "Our Best and Worst Presidents:  Some Possible Reasons for Perceived Performance."  Presidential Studies Quarterly 19: 529-557.

Holsti, Ole R.  1994.  "The Domestic and Foreign Policy Beliefs of American Leaders, During and After The Cold War."  Paper prepared for the Annual Meeting of the International Studies Association, Washington, D.C., 29 March-1 April.

Huntington, Samuel P.  1993.  "Clash of Civilizations?"  Foreign Affairs, no. 72 (Summer): 23-49.

Kane, Joseph Nathan.  1981.  Facts About American Presidents:  A Compilation of Biographical and Historical Information.  4th ed.  New York:  The H.W. Wilson Company.

Kegley, Charles W., Jr., and Gregory A. Raymond.  1992.  "Must We Fear a Post-Cold War Multipolar System?"  Journal of Conflict Resolution.  36, no. 3 (September): 573-85.

King, Gary and Lyn Ragsdale.  1988.  The Elusive Executive:  Discovering Statistical Patterns in the Presidency.  Washington, D.C.:  Congressional Quarterly Press.

Klingberg, Frank L.  1952.  "The Historical Alternation of Moods in American Foreign Policy."  World Politics 4 (January): 239-73.

------.  1983.  Cyclical Trends in American Foreign Policy Moods.  Lanham, Maryland: University Press of America, Inc.

------.  1996.  Positive Expectations of America's World Role:  Historical Cycles of Realistic Idealism.  Lanham, Maryland: University Press of America, Inc.

Kondratieff, N. D.  1926.  "The Long Wave in Economic Life."  Reprinted in The Review of Economic Statistics XVII (November 1935): 105-15.

Lembcke, Jerry.  1991-92.  "Why 50 Years?  Working-Class Formation and Long Economic Cycles."  Science and Society LV (Winter): 417-45.

Mager, Nathan H.  1987.  The Kondratieff Waves.  New York: Praeger.

McLoughlin, William G.  1978.  Revivals, Awakenings, and Reform:  An Essay on Religion and Social Change in America, 1607-1977.  Chicago, Illinois: University of Chicago Press.

Mensch, Gerhard.  1979.  Stalemate in Technology:  Innovations Overcome the Depression.  Cambridge, Massachusetts: Ballinger.

Modelski, George.  1981a.  "Long Cycles, Kondratieffs, and Alternating Innovations: Implications for U.S. Foreign Policy Behavior."  In The Political Economy of Foreign Policy Behavior, ed. C.W. Kegley and P. McGowen, 63-83.  Beverly Hills, California:  Sage Publications.

------.  1987a.  "The Study of Long Cycles."  In Exploring Long Cycles.  George Modelski, ed, 1-15.  Boulder, Colorado:  Lynne Rienner Publisher.

Modelski, George, and William R. Thompson.  1987.  "Testing Cobweb Models of the Long Cycle."   In Exploring Long Cycles, George Modelski, ed., 85-111.  Boulder, Colorado:  Lynne Rienner Publisher.

------.  1988.  Seapower in Global Politics, 1494-1993. Seattle, Washington:  University of Washington Press.

Namenwirth, J. Zvi.  1973.  "Wheels of Time and the Interdependence of Value Change in America."  Journal of Interdisciplinary History 3: 649-83.

Pollins, Brian M., and Randall L. Schweller.  1997.  The Long Wave and Mood Swings in U.S. Foreign Policy, 1792-1993.  Paper prepared for the Annual Meeting of the International Studies Association, Toronto, 18-22 March.

Rasler, K.A. and W. R. Thompson.  1994.  The Great Powers and Global Struggle.  Lexington, Kentucky:  University Press of Kentucky.

Reichley, A. James.  1994.  =93Get Ready for 60 Years of GOP Rule.=94  Wall Street Journal.  December 5:  A14.

Richman, Alvin.  1994.  "American Public's Attitudes Toward U.S. International Involvement in the Post-Cold War Era."  Paper prepared for delivery at the Annual Meeting of the International Studies Association, Washington, D.C., 30 March.

Schlesinger, Jr., Arthur.  1986.  The Cycles of American History.  Boston, Massachusetts: Houghton Mifflin Co.

Schlesinger, Sr., Arthur.  1949.  Paths to the Present.  New York:  Macmillan.

Singer, Max and Aaron Wildavsky.  1993.  The Real World Order:  Zones of Peace/Zones of Turmoil.  Chatham, New Jersey:  Chatham House Publishers, Inc.

Small, Melvin and J. David Singer.  1982.  Resort to Arms:  International and Civil Wars, 1816-1980.  Beverly Hills, California:  Sage Publications.

Statistical Abstract of the United States.  Annual.  Washington, D.C.: Government Printing Office.

Strauss, William, and Neil Howe.  1991.  Generations:  The History of America's Future, 1584 to 2069.  New York:  William Morrow and Company.

Thompson, William R.  1988.  On Global War:  Historical-Structural Approaches to World Politics.  Columbia, South Carolina:  University of South Carolina Press.

van Duijn, J.J.  1981.  "Fluctuations in Innovations Over Time."  Futures 13 (August): 264-75.

Wilkinson, D.  1992.  "Decline Phases in Civilizations, Regions, and Oikumes."  Paper presented at the Annual Convention of the International Studies Association, Atlanta, 2 April.