2
second image theory of hierarchy that systematically addresses when dominant states will
successfully manage their affairs so that they retain their status and when they will fail.
Committing to self- restraint is difficult. Such commitments are most credible
when political groups within the state with a disproportionate interest in the stability of
the working relationship with the subordinate state hold a veto over foreign policy
decisions in the dominant state. As an initial exploration of the problem, I contrast two
historical episodes in the development of the Empire. In the first the British failed to
exercise self-restraint in the governance of the American Colonies and as a consequence
they lost control of North America. In the second instance a veto group with specific
investments of the kind I describe existed in London, creating an expectation among
Australians that Britain would be consistently restrained. As a consequence, Britain did
not face a revolt from the Australian Colonies in a situation that was otherwise similar to
the events leading up to 1776.
The paper proceeds in three parts. First, I outline a need for a second image theory
of self-restraining behavior in hierarchic relationships. Second, I develop an argument
about the conditions in which hierarchic states will make self-restraint credible. Third, I
contrast the history of the British relationship with the American Colonies in the 1760s
with the British relationship with the Australian Colonies in the 1870s.
1.
The need for a second image theory of hierarchy
Just because a strong state could be better off in the long run by practicing self-
restraint, there is no guarantee that it actually will. As many international relations
scholars have noted, empires often overreach to the point of self-destruction. A systemic
theory is unlikely to be able to account for why some empires have been unable to
restrain themselves and thereby inspired rebellion (such as the Soviet Union or the first
British Empire) and some have been able to commit to moderation (the second British
Empire). Second image theories, which explain differences in outcomes by referring to
differences in the internal workings of states, have that potential.
Knowing why different kinds of states in otherwise similar positions might be
more or less successful at imperial management is important for theoretical reasons that
go beyond just understanding the persistence or collapse of empires. A second- image
theory of self-restraint is critical to any complete theory of unipolarity.
Balancing behavior is to multipolarity as self- restraint is to unipolarity.
1
Both are
unit- level behaviors that are critical to systemic outcomes. An important difference,
however, concerns the importance for systemic stability of decisions made by particular
units (states). Consider a multipolar system. Neorealist, as well as most traditional
realist, theories predict that multipolar systems will produce balancing behavior.
Kenneth Waltz’ cornerstone theory predicts that in multipolar systems balances will tend
to form, even though the theory does not predict – and does not need to predict – the
behavior of any one state (Waltz 1979). If one state or coalition of states grows more
powerful, enough other states will flock to the weaker side to restore a balance of power,
so that even if some states engage in the out-of-equilibrium behavior of flocking to the
stronger side the overall balance is nevertheless restored. Balancing is the critical
1
By “balancing behavior” I am referring to the unit -level choice, not the systemic outcome. A state
engages in balancing behavior when, during a period of general tension or war, it joins the weaker side.
Convention