Review Essay 47 on Twilight of the Titans: Great Power Decline and Retrenchment17 min read

Paul MacDonald and Joseph Parent bring to book-length form a very sensible and persuasive argument that they have been making for some time. Great power decline is not necessarily dangerous or even destabilizing. Countries can pursue strategies of retrenchment, either of “self-help” by cutting back spending or rejuvenating their economy, or of external adjustment in paring back commitments or cementing new friendships. Such strategies, MacDonald and Parent argue, need not be destabilizing. The countries experiencing decline can regain strength and confidence.

H-Diplo | ISSF Review Essay 47

Paul K. MacDonald and Joseph M. Parent.  Twilight of the Titans: Great Power Decline and Retrenchment.  Ithaca:  Cornell University Press, 2018.  ISBN:  978-1-5017-1709-3 (hardcover, $42.95).

Reviewed by Philip Zelikow, University of Virginia

Published 1 February 2019|

Editor: Diane Labrosse
Web and Production Editor:  George Fujii



Decline or Renewal?

Paul MacDonald and Joseph Parent bring to book-length form a very sensible and persuasive argument that they have been making for some time. Great power decline is not necessarily dangerous or even destabilizing. Countries can pursue strategies of retrenchment, either of “self-help” by cutting back spending or rejuvenating their economy, or of external adjustment in paring back commitments or cementing new friendships. Such strategies, MacDonald and Parent argue, need not be destabilizing. The countries experiencing decline can regain strength and confidence.

MacDonald and Parent first published this argument at length in 2011, in both International Security (the lead article in that issue) and Foreign Affairs.[1] It seemed timely then. America was going through its post-Iraq/Afghanistan phase of declinism. The book certainly seems timely now, in 2018, since Americans are going through another phase of declinism, this one mainly focused on China, piled on to other causes for gloom, like the antics of the current American president.

Recent scares, for instance like those about China of Graham Allison with his argument about a supposedly timeless “Thucydides trap,” are overdrawn and actually dangerous.[2] They talk of war as a probable outcome, when it rarely is. Thus such alarms are, the authors imply, a bit like crying, “Fire!” in a crowded theater instead of just attending to the offending cigarette butt.

When MacDonald and Parent first made their case, back in 2011, the retrenchment argument seemed to play into a zeitgeist hoping for a return to “offshore balancing,” a strategic concept offered by other theorists classified as neo-realists. At the time MacDonald and Parent seemed to accept that situation, but they were open to other forms of retrenchment. Now their book is especially agnostic about how to do the retrenchment. The book’s principal, and most valuable, addition to the earlier articles is a set of six historical case studies, involving Britain, France, and Russia between the 1870s and the 1920s. The case studies illustrate a rich menu of strategies of retrenchment, and the authors are plainly open to whatever works under the circumstances of the times.

MacDonald and Parent are especially anxious not to align themselves with either side in the current fashionable binary of ‘liberal internationalists’ versus ‘realist retrenchers.’ They stress that their notion of retrenchment is not necessarily a retreat at all. A principal reason for offering their historical cases is to refute such a shallow binary choice. So they expressly distance themselves from the “repudiation of American dominance in international life” (198, referring to John Mearsheimer and Barry Posen) which they associate with some neorealists. Since retrenchment does not necessarily require an abandonment of America’s position in the world (whatever that means, case by case), they also wave off the fears of the internationalists, who worry that, “Retrenchment would in essence entail a massive [and dangerous] experiment” (196, citing Stephen Brooks, John Ikenberry, and William Wohlforth).[3]

MacDonald and Parent are trying to replace a cartoon image of retrenchment with more complex illustrations of the reality. They hope that practical people, like them, will take charge and adopt policies that are well-judged, pulling back some forces here, but not there, moving spending around more effectively, adjusting policies both at home and abroad. They observe, nicely, that, “retrenchment is not the devil great powers don’t know, but the devil they know best. Decline happens commonly and retrenchment is the most common response to it” (196).

“Decline happens commonly.” That sure is right. Take just the United States. Writing in 1988, Samuel Huntington observed that, “In 1988 the United States reached the zenith of its fifth wave of declinism since the 1950s.”[4] And I count at least two before the 1950s and at least one or two more such waves in the twenty-first century. MacDonald and Parent cite Walter Russell Mead (in 2014) proclaiming American decline amid a “return of geopolitics” (179). Huntington, back in 1988, cited the same Walter Russell Mead (in 1987) mourning American decline in Mortal Splendor. Mead published that book in the same year Donald Trump started taking out big newspaper ads denouncing the passing of American greatness. That was about the time a former U.S. official announced that, “The American Century is over. The big development in the latter part of the century is the emergence of Japan as a major superpower.”[5]

Huntington’s foresight looks a little better: he thought this particular bout of declinism was actually renewal. His article came out in the winter of 1988-89. The next few years were kind to his argument.

The frequent ups and downs do raise a rather awkward problem for part of MacDonald and Parent’s argument. Like their original International Security article, the first section of the book relies on a supposed dataset for all cases of great power decline from 1870 to 2018. There were 18 of these in the 2011 article, 16 in the current dataset in which not one of the declines happened to the United States. To them, the U.S., and China (which they classify as a great power since 1950), “have not experienced decline” (46).

This will come as news to students of Chinese history who recall the effects of the Great Leap Forward or of the Great Proletarian Cultural Revolution. It will also obviously be news to any student of the United States. For instance, when Daniel Sargent published his justifiably acclaimed recent book, A Superpower Transformed, about “the remaking of American Foreign Relations in the 1970s,” he was not describing a transformation because of an advance in America’s relative power.[6]

MacDonald and Parent get into this and other trouble with their dataset because it relies entirely on relative shifts in great power GDP (including historical GDP as measured by British economist Angus Maddison). They have heard this critique before, including in one of the replies to their 2011 article. Their answer then, and now, is that the GDP metric is objective, that the alternative indices of decline cannot be “generalized across time and space.” In any case, they write, the critics agree with the underlying argument.[7] As do I.

I would, however, have preferred an avowedly subjective dataset. Look at each of the histories of the nine countries they classify as great powers during their timeframe. In each of those histories, try to identify the times when those countries had a great debate about relative decline, including the contemplation or implementation of alternative grand strategies for revival or retrenchment (or preventive war, which was considered in the U.S. between 1949 and 1954).

Such a dataset could be challenged, of course, because of all the subjective coding. Yet MacDonald and Parent’s preference for a supposedly objective metric that can be “generalized across time and space” presents a more profound philosophical problem. What are they cataloguing? Great power “decline” is a social reality. It is a compound of subjective beliefs about circumstances, in which even the circumstances themselves are a mix of ‘brute facts’ along with presumptions and conjectures.[8] If one is arguing about how people react to beliefs about decline, the epistemology of the data must align with the epistemology of what is being studied.

MacDonald and Parent assert that “economic data … are likely to correspond with crises” (45). Since their particular metric missed both of the giant China cases, there are obviously some problems with the way it picks up relevant economic changes. But in general I do not believe their observation on this point is accurate. The Boer War funk, Sputnik/Berlin, 9/11-Iraq and many more instances come to mind.

Yet the good news for MacDonald and Parent is that if they followed my subjective, historicist advice and if, as a result, their dataset trebled in size, the results would still reinforce their underlying argument. There would be that many more examples of the varieties of behavior, including constructive retrenchment. Perhaps the old dataset was included simply to make one kind of disciplinary audience happier.

It is the six historical cases (Britain, France, and Russia from 1870s to 1920s) that are the major research additions in the book. These are quite well done. Although the authors seem to work only in secondary sources in English, they have combed that literature very thoroughly and their summaries seem sound.

The most important insight from the cases is that retrenchment strategies tend to be complex and multifaceted. Recall that in the IR-world fight between the cartoons (parsimonious depictions?) of ‘offshore’ retreaters and globalist internationalist blobs, MacDonald and Parent insist to both of these camps that ‘retrenchment’ did not have to fit either of these caricatures. Their historical cases prove that point.

Had they discussed two periods of Chinese worry about relative decline, the results would have added still more nuance, or just contingency.

During the Great Leap Forward catastrophe (1958-1962), while tens of millions of Chinese were starving and rapid industrialization was a disaster, China’s relative power versus the United States was presumably declining. The Chinese government became more aggressive—finalizing the breach with the Soviet Union and denouncing Moscow; attacking India; and eventually adding a major troop commitment to back North Vietnam’s moves south, along with a threat of war against America if U.S. troops moved into the north.

As a result of the Cultural Revolution implosion, which killed one or two million Chinese and profoundly destabilized the entire society, China’s relative power compared to the USSR certainly declined. China’s initial development of nuclear weapons was not necessarily a stabilizing factor. This time the Chinese government engaged in a definite grand strategy of retrenchment. This strategy included a rapprochement with the United States, initiated as least as much by China as by the Nixon administration.

One could argue that both sets of strategies were effective, by Chairman Mao Zedong’s lights at least.

Similarly, had the authors discussed the American response to decline during the 1970s, they would have described strategies of retrenchment of just the mixed, calculated kind they wish to spotlight. It is not entirely clear that the dominant theme is retrenchment. Even in East Asia, these were not strategies of wholesale retreat. President Carter’s initial impulse to withdraw from South Korea was checked. Meanwhile, in the Middle East, U.S. policies were more activist, not less.

At the end of the 1970s and beginning of the 1980s, the U.S. perception of continued decline led to a set of new policies in domestic and global finance, in general rearmament, and in military posture in Western Europe. But it would not be quite fair to call these an aggressive expansion of American commitments. The two most active efforts to extend commitments, in Lebanon and in Central America, did not fare well. New commitments to Persian Gulf security were more modest and useful.

The core work to rebalance commitments with means took place in Europe and East Asia, and it met with a good deal of success in both regions. So was this another kind of retrenchment? Contemporaries did not think so. Associated with the booming budget and trade deficits of the second term of the Reagan presidency, these exertions triggered the bout of late 1980s declinism that so drew Huntington’s ire.

Also, to break the authors’ purely national construct, during the late 1970s the American polity was divided on both the global finance and the military posture moves. In both cases, the strategic initiative was pressed on America in 1978 and 1979 by Europeans. They were led in particular by the West Germans, using the new G-6/G-7, the new European Monetary System, and the institutions of NATO. Chancellor Helmut Schmidt for example (supported by French President Valéry Giscard d’Estaing), was also reacting to beliefs about relative decline, but not just theirs.[9]

Once retrenchment escapes the cartoon image of retreat, labels become difficult. When the George H.W. Bush administration took office in 1989, it sought to counter worries, then rife, about American decline. It developed an affirmative agenda as the Cold War also ended. From this point of view, the Uruguay Round and creation of the World Trade Organization (WTO) might be deemed a strategy of retrenchment, a response to decline and the collapsing General Agreement on Tariffs and Trade (GATT), but the strategy certainly did not adopt a stance of retreat. In security matters, the Bush administration undertook the Gulf war while it was also beginning a program of large-scale cuts in defense spending, a big part of a general fiscal retrenchment. The Bush administration’s approach to NATO was to preserve the coupling of America’s security with that of Europe (a coupling many thought was now outmoded), while transforming the nature of the alliance from a primarily military coalition to a primarily political organization mainly worrying about post-Communist transition. This transformation was accompanied by large-scale arms control and force reductions. Yet few people think of this as a strategy of “retrenchment” in Europe.

To me, one of the most interesting possibilities that emerges from the authors’ cases is that countries undertaking “retrenchment” can be more endangered by their success than by their decline. Reverses can spur renewal. But success can lead to ‘victory disease.’

Consider the account the authors give of Russia in 1903. The authors pick 1903 for the great year of decline, but they could just as well have chosen 1904-1905, years of military defeat and internal revolution. All the world, and all Russians, saw great decline.

What did the Russians do in this time of desperate failure? They regrouped and, as MacDonald and Parent recount, engineered a remarkable and rapid revival of economic, fiscal, and military power. They stepped back from their ambitions in East Asia, patched up a generations-old rivalry with Britain, and shored up their alliance with France. Russia became the fourth largest economy in the world, its national income 97% that of Britain. By 1912 the transformation astonished Europe.

At the very end of their case, MacDonald and Parent look at how Russia reacted to this remarkable renewal. In 1912 Russia embarked on its “Big Program” to build up its armed forces, a program set to culminate quite formidably in 1917. MacDonald and Parent regard this as a case where, “Retrenchment worked, but in the end a combination of new circumstances and old leadership meant that reforms neither went far enough nor endured long enough” (161).

That does not go quite far enough. MacDonald and Parent are right that the Russian leadership after 1912 became both assertive and incompetent, a rather volatile mix (160). “Retrenchment worked,” indeed. Russia had felt that it had had to back down in one Balkan crisis after another from 1908 to 1913. Now, with successful retrenchment, Russian leaders felt they could get tough the next time.

Because it did not fit their GDP metric, MacDonald and Parent leave out of their dataset how deeply Germany noticed its relative and impending decline.

On the evening of 7 July 1914, having just authorized a blank check to Austria-Hungary to take strong measures in the crisis over the assassination of its Archduke, the German Chancellor, Theobald von Bethmann Hollweg, was out at his Hohenfinow estate, talking with his young aide, Kurt Riezler. Riezler noted in his diary:

The old chateau, the wonderful enormous trees, the avenue like a vaulted gothic ceiling. … Melancholy and restraint in landscape and people.

In the evening on the verandah, under the night sky, long conversation of the situation … Russia’s military power growing rapidly; following strategic build-up of [Russian] Poland situation untenable; Austria continually weakening and more immobile; the undermining [of Austria-Hungary] from North [Russian Poland], South East [Balkans] far advanced.[10]

Bethmann was not planning a preventive war. At that point he and his Kaiser and his colleagues were just trying to bolster their ally, not very thoughtfully.

But part of the way the crisis developed was a product as much of Russia’s successful renewal as of the German fears this triggered. This time, in this Balkan crisis, the attitude of Russian ministers was “characterized by arrogance and anxiety in equal measure.”[11]

Arguments about decline are common. Often overdrawn, such arguments are healthy in a great power if they don’t turn into full-blown hypochondria. The times of success are often more dangerous. The U.S. defense establishment was more innovative in preparing for future conflict in the 1970s and 1980s, after Vietnam and after it thought its military posture in Europe was hollow, than it was in the 1990s, after the Gulf war victory of 1991. It was after it had succeeded with graduated escalation in the Cuban Missile Crisis that the U.S. tried graduated escalation in Vietnam. It was after it had succeeded with surprising ease (at first) in Afghanistan that the U.S. ran headlong into Iraq in 2002-03. Lots of things can be worse than worries about decline.


Philip Zelikow is the White Burkett Miller Professor of History and J. Wilson Newman Professor of Governance at the Miller Center of Public Affairs, at the University of Virginia. He has worked at all levels of American government, including in each of the five administrations from Reagan through Obama.

©2018 The Authors | Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License



[1] Paul MacDonald and Joseph Parent, “Graceful Decline? The Surprising Success of Great Power Retrenchment,” International Security 35:4 (April 2011): 7-44; Joseph Parent and Paul MacDonald, “The Wisdom of Retrenchment: America Must Cut Back to Move Forward,” Foreign Affairs 90:6 (November 2011): 32-47.

[2] Graham Allison, Destined for War: Can America and China Escape Thucydides’s Trap? (New York: Houghton Mifflin Harcourt, 2017).

[3] For the quotation, see Stephen Brooks, G. John Ikenberry, & William Wohlforth, “Don’t Come Home, America: The Case against Retrenchment,” International Security 37:3 (Winter 2012/13): 7, 10.

[4] Samuel P. Huntington, “The U.S.–Decline or Renewal?,” Foreign Affairs 67:2 (Winter 1988/89): 76-92.

[5] Huntington cited Mead’s Mortal Splendor: The American Empire in Transition (Boston: Houghton Mifflin, 1987),remarking on American decline since the 1960s, one of a cluster of books echoing that theme at the time. MacDonald and Parent cite Mead’s, “The Return of Geopolitics: The Revenge of the Revisionist Powers,” Foreign Affairs 93:3 (May-June 2014): 69-79, although this essay’s argument is not actually as declinist as that in the earlier book. “The American Century is over …” is from Huntington citing Clyde Prestowitz in Time, July 4, 1988, 28.

[6] Daniel Sargent, A Superpower Transformed: The Remaking of American Foreign Relations in the 1970s (New York: Oxford University Press, 2015).

[7] See especially the letter from Kyle Haynes and the authors’ replies in “Correspondence: Decline and Retrenchment – Peril or Promise,” International Security 3:4 (April 2012): 189-203. Haynes was a doctoral student at my university, the University of Virginia, but (sadly) I did not know him.

[8] Territory and population are brute facts that exist independent of our agreement (although borders are a more complex case). GDP is an institutional fact, a social construct in which people believe or not, but which may be a proxy for an external reality of material conditions. Widespread belief in a decline in relative power is a much more contestable social construction, which presupposes certain external realities, including about current material conditions, but goes well beyond them. Beliefs about “decline” include presumptions and conjectures about ourselves, about others, and about the objects and utility of the supposed powers. An outsider can observe the existence and apparent influence of such beliefs, on which people often act regardless of whether or not some outsider or later expert regards those beliefs as being true. See John Searle, The Construction of Social Reality (New York: Simon & Schuster, 1995).

[9] For a brief version of the argument, see Kristina Spohr, The Global Chancellor: Helmut Schmidt and the Reshaping of the International Order (New York: Oxford University Press, 2016).

[10] Riezler quoted in T.G. Otte, July Crisis: The World’s Descent into War, Summer 1914 (Cambridge: Cambridge University Press, 2014), 92.

[11] Otte, July Crisis, 122.