One of the oldest questions in political and philosophical debates is that of when one state ends and another begins. In The Histories, Herodotus tells us about a meeting of Greek commanders during the second Persian war to decide upon their naval strategies against the Persian fleet, in the wake of the Persian breakthrough on land—Athens had been sacked, and the surviving Athenians evacuated to Salamis. In the middle of the meeting, the Athenian commander, Themistocles, is told by Adeimantus, the Corinthian representative, to keep quiet, when trying to persuade the united Greek fleet to fight the Persians at Salamis. Adeimantus says, ‘Themistocles should find himself a state before contributing his ideas’, as Athens was fallen and occupied. Themistocles rebukes the idea, maintaining instead Athenians still have a city as long as they have two hundred ships ready for action. Later, Themistocles threatens the other Greek commanders that if they do not stay at Salamis to fight, the Athenian fleet will sail away to Italy and settle their city-state there, leaving the other city-states to deal with the Persians on their own.
We know that eventually didn’t happen. The Athenians stayed and together with other Greeks expelled the Persians. Thus, it seems no longer a serious question to ask whether Athens would truly continue to exist if the original city was never restored while the survivors claimed to relocate the city-state in Italy. But if we situate ourselves in the earlier moment when the Athenians had to decide what they ought to do in relation to their lost city, this question cannot be set aside lightly, especially if the continuity of Athens as one persisting city-state is what matters for them. Should they—we—stay to fight to restore the city to save the city-state? Or can we continue the city-state just in a different place, with a new city?
This exemplifies the problem of the continuity of states. Continuity in this context means identity over time, and identity as such means numerical identity or quantitative identity, meaning that there is one, and only one thing which has existed there for the given period of time. A different but related sense of identity is qualitative identity; it reflects the sense of sameness we use when we utter such sentences as you and I bought the same book. There are two books, and completely alike.
The continuity of states is primarily concerned with the first sense of identity, for it asks whether it is the same state which persists over the given period of time. But the second sense of identity is often relevant to the analysis too. On the one hand, numerical identity over time may depend on qualitative identity being maintained in a sufficient degree. On the other, it is not unthinkable that under some conditions we cannot get a final, determinate answer about whether a situation is one state which persists, or two successive states which are immensely alike.
For example, if a new Athens is created in Italy, with all the old Athens’s key political, legal, and social features faithfully replicated, we may come to a point where we can either say it is the same Athens that continues to exist, or a different city-state which comes into existence and is just like Athens. And neither is true or false. They are only different, alternative descriptions of the same state of affairs.
The nature of the state is essential to a full analysis of continuity. The Themistocles-Adeimantus debate reveals two ways of understanding a state, viz. as a group of people and as a piece of land. Aristotle introduces a third: as a constitution. It is his famous thesis that the form of the city-state’s constitution determines whether a city-state continues to exist as the same city-state. The people can be totally different without changing the city-state’s identity. Likewise, the city-state’s identity can be ruptured due to constitutional break while all people are the same. Scholarly debates to date can be very roughly seen as one among these lines of thinking, among modified, more sophisticated versions of each understanding of the state.
I would like to show that the truth is likely to lie in not any one, but the combination of some (and not all) of the lines of thinking. My strategy is to look at each major component of the state such as people, space, power, and over time, culture, and explain how their continuation and evolution over time might bear on the identity of the whole unit whose components they are.
The analysis is pertinent to an important question: how important is political self-determination to a state’s continuity? A national community has an understanding about who it is, or which historical community it is continuous with. I call this the community’s cross-temporal self-identification. We tend to think this should play a key role in the community’s continuing identity. But further analysis may indicate a much more qualified role. What I think I can say now is that a community’s self-identification is a cause of its continuity, but the identification does not in itself translate into the actual continuity of the community. Sometimes, in fact, the self-identification can be blatantly false. Just as a person is not always whom they think they are, so too a community cannot be whatever it identifies itself with. In other words, we are not always what we believe we are.