Keywords

1 Introduction

The notion of explanation is fundamental in both scientific theories and philosophical accounts of how science works. A number of chapters elsewhere in this volume will address the issue of explanation in historiography and evolutionary science from a philosophical perspective, but this chapter will look at the topic in a complementary way: Debates from current historiography of ancient Rome will be presented as a case study. This is intended to elucidate some of the problems that historiographers actually encounter in their work, i.e., problems that can be observed particularly in cases where unequivocally accepted explanations are absent.

In principle, problems of methodology are similar across all disciplines that deal with history in a broad sense of the word: history, archaeology, linguistic science, and so on in the humanities but also paleontology, evolutionary biology, or cosmology in the natural sciences. This fundamental similarity is a basic tenet of several works elsewhere in this issue, and the author of this chapter has also argued in favor of this methodological similarity (Berry 1999, 2008). However, similar problems and approaches do not imply identity in all respects. Human actors have attributes, such as intentions, plans, and beliefs, which are usually absent from the objects of study in historical fields of natural science: the emergence of our solar system can be reconstructed without any reference to the intentions of the planets, while mental states are essential for understanding events in human history.

This gives rise to a special situation in historiography: apparently plausible explanations for historical phenomena are usually rather easy to present because the modern observer feels related to the actors of the past through concepts such as feelings, needs, intentions, and the like. The fundamental similarity between observer and observed enables the construction of seemingly plausible ad hoc explanations in many cases: actor X did Y because he wanted to achieve Z (see, for instance, the habit of archaeologists to ascribe a “religious” or “ritual” purpose to any object for which other functions are not obvious).

Completely unexpected and unexplainable phenomena probably will be rare in human history, but occur rather often in the fields of natural history. When in 1995 the first planets outside our solar system were detected, they exhibited a number of features that were unexpected and unaccounted for in astrophysical theory. The existence of exoplanets as such had actually been anticipated. But their large atmosphere in combination with the small distance to their sun was not merely unexpected; it was something that one would have considered as outright impossible prior to this discovery: intense heat and radiation from the parent star should have blown all remnants of an atmosphere away, according to generalizations from observations of our own solar system. By contrast, the major problem that historians and archaeologists frequently face is to select between several equally conceivable explanations; the complete absence of explanations, however, is less frequent.

To start the tour through Roman history, the next section will take a look at the available sources. In addition to problems of interpretation, with which anybody working with documents from the past is confronted, Roman sources have peculiar features that affect the problem of finding explanations for some aspects of Roman history.

We will then turn to two major questions: First, the rise of Republican Rome from a city to a world empire. How can we explain that a single city became a superpower that could dominate the whole Mediterranean basin and many adjacent territories? Second, the inner workings of the Empire once it had conquered the world: how can we explain the decision-making process in the center of power? The logical third step would be the decline and collapse of the Empire. This is the mystery of mysteries in history, and countless explanations have been offered. Therefore, this can of worms will not be opened here, given the limitations of space (see Ando 2008; Wolfram 1990, pp. 422–441; Demandt 1989, pp. 470–492 for an overview).

2 The Problems of the Sources

Rome is one of the most intensely investigated cultures and states of the past. The amount of things that we know about ancient Rome is enormous. Nevertheless, some areas are particularly prone to produce long-standing controversies and, correspondingly, a lack of undisputed explanations. So what are the gaps in the available evidence, leaving questions of Roman history unanswered, despite centuries of scholarly work?

There are two main types of evidence: material remains, as studied by archaeologists or art historians, and written accounts, studied by historians and philologists. In between, there are categories such as papyri, inscriptions on buildings, and coins, which fall in both realms because they are material objects on the one hand but contain textual information on the other.

Material evidence can tell us a great number of things about ancient conditions of life, of trade routes, of production processes, and so on. But with respect to causal explanations, in particular when it comes to explaining political or social processes, material evidence has limits. Let us take as an example a Roman glass vessel that is found in Germania Magna, the unoccupied part of Germany on the right-hand side of the Rhine. Chemical analyses will reveal the composition of the glass, and by some fancy methods, it may be possible to trace the provenience of the raw materials that were used, and perhaps one can even locate the workshop where the glass was made.

But how did it come into the soil in Germania, perhaps hundreds of miles away from the Roman limes? Conceivable explanations could be:

  • It reached the barbaricum by means of normal trade.

  • It was loot that Germanic raiders of the Roman Empire had brought home.

  • It was given by the Romans to some Germanic king or chieftain, as part of diplomatic exchange of gifts.

  • It was a piece that a Germanic mercenary in Roman service had acquired and brought home, when he returned after his term of service.

This example is intended to show that, using material evidence, we can answer many questions regarding how, when, and where the people of the past did what they did, but the central questions “why?” and “in which historical context?” are generally more difficult or impossible to answer, based on material evidence alone.

Answers to such questions concerning the intentions and motivations of the players in history are more aptly sought in the historical accounts, but these have their limits, too.

First, not everything that has happened was captured in written form, and, second, not everything that was written down has survived to the present day. These are trivial problems that relate to any written account of the past. Likewise, writing history was a pastime for members of the upper classes, giving rise to a considerable social bias in their writings, but again this is a problem that we are frequently confronted with in any historiography of the premodern age. Some more specific problems that relate especially to Roman history are depicted in the following section.

2.1 The Classical Model

Ancient historiography of later eras, i.e., Hellenistic Greece as well as Republican and Imperial Rome, took the earlier works of Classical Greece to be an authoritative model that was to be emulated as far as possible. This gave rise to an approach that squeezed the material into a canonical form, irrespective of whether this did justice to the matter at hand or not.

And historiography had a number of different purposes: to educate, entertain or surprise the reader, or to make a political point, rather than to capture the course of history in an objective or scientific way. The problem that the ancient sources deliberately blur the picture and that an elegant reading and the adherence to the canonical pattern is more important than accurate detail can affect any context that is “technical” in the broadest sense: technology proper, military matters, economic, administrative, or legal affairs. And then there were other topics which were not deliberately blurred but which were simply too trivial and too self-explanatory for the ancient reader to be expounded explicitly.

2.2 Rome as the Center of the Universe

For Roman historians, the city of Rome was the center of the universe, and “Roman history” was the history of that city. The empire-wide effects of Roman rule and decisions were not relevant, at least not in themselves. This means that issues which are important for modern historians, such as social and economic history, have to be reconstructed in a tiresome way from pieces of scattered evidence, often archaeological or epigraphic in nature, because the writings of ancient historians offer only meager material on these questions. In general, the written accounts are either Roman or Greek in perspective, and other people living under Roman rule remain silent for us.

The only and notable exception are the Jews because a number of sources allow to see the Empire with their eyes, in particular the works of Flavius Josephus. He is essentially the only author who follows the patterns of Greek and Roman historiography but writes from the perspective of a people on the periphery of the Greco-Roman world. Due to a lack of comparable sources, it is impossible for us to complement this with a national history of, say, the life of Numidian or Illyrian tribes under the empire.

2.3 The Not So Impartial Observer

Many Roman historians of the imperial time were senators which means that they, in addition to the mentioned general upper-class bias, also had a marked anti-emperor bias because the relations between emperor and the senate were frequently strained. This caused senatorial historians to use their writings for revenge, usually after the emperor in question was dead.

2.4 Politics in Secret

While politics in the republic had been at least partially a public affair, being discussed in the senate, the forum and the people’s assembly, it had become something essentially secret under the emperors, with decisions being made in the inner circle of power. Therefore, in the imperial era, historians were able to properly expound the backgrounds and causes of political decisions only to a limited extent – and this limitation has, of course, been inherited by their modern successors. And then there are examples of ancient historians who actually belonged to these inner circles at some point of their career. Yet this does not guarantee that their accounts are particularly reliable, because of their involvement in court intrigue and the urge to use their knowledge for retaliation, as described above.

The previously mentioned problems relate to the interpretation of available sources, but with regard to the earliest phases of Rome, we face an additional difficulty: there were no contemporary writers, so that all material on the times of the kings, the origin of the republic, and the beginnings of its rise to dominant power in Italy originated at a later date. Now it is time to look at the rise of Rome as a superpower.

3 The Rise of Rome

Already in ancient times, observers were bothered by the question how a single city came to dominate the whole Mediterranean world. The Greek Polybius, who wrote in the second century BC, was the first historiographer to tackle this issue in a systematic manner. His approach, i.e., asking for the cause of the rise of Rome, stands within a tradition of causal analysis that goes back to some of the most important Greek historians: Herodotus, the so-called founder of historiography, was interested in the cause of the wars between Persians and Greeks (the intricate story in Herodotus 5.23–97 includes earlier conflicts within the Greek world that preceded the Ionian Revolt against the Persians). Likewise, Thucydides wrote his work to elucidate the cause of the Peloponnesian War between Athens and Sparta (immediate causes of the war, Thucydides 1.23–87; the underlying long-term conflict between Athens and Sparta, 1.88–118).

One causal factor that Polybius regards as crucial is the constitution of Rome, which, according to him, is a perfect balance of different types of constitution known from the Greek cities (Polybius 6.18; comparison of Sparta’s and Rome’s ability to create a stable hegemony, 6.48–50). In addition, he considers the Roman army to be a crucial factor (6.19–42); this type of reasoning is popular even today: the Romans dominated the world because they had the best army in the world.

One will not encounter this explanation in academic circles today, but in the mass media, in accounts for a general audience, it is still alive. But what does “best” army in the world mean? – This is an undefined term. Moreover, here it seems appropriate to make a direct comparison to the present age; having the best army in the world is not enough: the US forces may be called the best army of the present, but winning battles is not the same as winning wars or creating a stable peace order, as the situation in Iraq or Afghanistan shows so clearly.

For Polybius, it was in particular the tactical superiority of the more flexible Roman legion over the rigid Macedonian phalanx, which had become a standard formation for many powers around the Mediterranean (Polybius 18.31 f.). This explanation of Roman success is also found in the Roman historian Livy, writing in the time of Augustus, i.e., about 150 years after Polybius (Livy 44.40–42). But looking at the encounters of legio and phalanx in detail, one sees that several times the Romans avoided defeat only by a hair’s breadth, so the notion of a general superiority cannot be maintained (analyses of such battles in Cowan 2009, pp. 103–147;Pietrykowski 2009, pp. 195–236).

In general, Roman history was full of severe defeats, what calls the whole approach to this explanation into question. In spite of the undeniable qualities of the Roman army, one has to conclude: Rome was not invincible, and the crucial feature that requires an explanation is the resilience of the Roman state, i.e., the ability to create a military and political system that remained intact even in the face of a total disaster, such as the catastrophic defeats against Hannibal’s army in the Second Punic War.

At this point, we must turn to the debate among modern historians. Their discussions of Roman expansion are centered on the notion of imperialism, which reveals that it originates in modern political science. In 1979, William Harris published “War and Imperialism in Republican Rome,” which became one of the most influential books on Roman expansion. At the beginning of Chap. 1, he states:

Since the Romans acquired their empire largely by fighting, we should investigate their attitudes towards war. (Harris 1979, p. 9)

In fact this is, in a nutshell, Harris’ program: he investigates the Roman mentality and concludes that it is the extraordinarily warlike character of the Romans that compelled them to uninterrupted warfare, year after year, for centuries, until they had finally conquered anything that was worth to be conquered.

In his review of the debate, Rich (2004) identifies two major reasons why Harris’ views became dominant: First, his theory replaced the older theory of “defensive imperialism,” which had been promoted by Mommsen and had become widely accepted. According to that view, the Romans only went to war because they had to; they were fighting essentially defensive wars during which, as a side effect, their empire constantly grew. This paradoxical view of unintentional world conquest had become untenable, and Harris’ approach seemed to offer a much more plausible explanation of the rise of Rome.

Second, the time at which Harris conceived his book was thoroughly influenced by modern anti-imperialism. The European powers had already lost their colonial empires, and exposing the evils of Roman imperialism clearly hit a nerve with many readers. The personal setting of Harris, who was writing as an Englishman in the United States during the Vietnam War, may explain in part the polemical character of his book, as Rich believes (see also Fitzpatrick 2010 on the comparison of ancient and modern “imperialism”).

Harris’ theory of a specific Roman urge to go to war had two facets. One is the immediate material benefits of conquest, i.e., the increase of territory, the influx of loot and money from plundering cities and selling their population as slaves, and so on. Concerning this aspect of the theory, Erich Gruen has demonstrated that the crucial decisions of the senate, when and where to go to war, were not generally dominated by economic motives:

A growing body of scholarly literature finds war and greed tantamount to imperialism. The equation may be too simple. Distinctions need to be made and emphasized. The prospect of loot could entice generals and stimulate recruiting – which is not the same as determining a senatorial decision to make war. The carrying off of spoils and the exaction of indemnity might enrich the state, but would not necessarily impel it toward an enduring system of regulation and exploitation. Enslavement or sale of defeated enemies helped stock the plantations of rural Italy; yet nothing shows that this either inspired Roman expansion or dictated imperial control. The leaps of logic too easily distort and mislead. (Gruen 2004, p. 30)

But mere greed, the drive for material rewards, would at least have had some rational core. According to Harris, there is another, even darker, and wholly irrational side of the Roman attitude. According to him, the Romans overrated warrior ethos and military glory to such a degree that their attitude became outright pathological. And it is because of the focus on the notion of a pathological Roman lust for war that Harris’ view has become popular.

Tim Cornell summarizes this standard view of Roman militarism in his comprehensive study of Rome’s early history:

For most of its history the Roman Republic was constantly at war, and a very high proportion of its citizen manpower was committed to military service. Its institutions were military in character and function, and its culture was pervaded by a warlike ethos. (Cornell 1995, p. 365)

However, he then introduces a new turn to the story, because he goes on:

These facts are important, but they do not explain Roman imperialism; rather, they are themselves symptoms of the phenomenon that needs to be explained. Why were the Romans so belligerent? How did they manage to conquer Italy so quickly, and why was their control of the conquered peoples so thorough and long-lasting? In the last analysis, the answer to all these questions is the same, and is to be found in the nature of Rome’s relations with her neighbors from the earliest times.

The foundations of Roman military power were firmly laid in the settlement that followed the Latin revolt in 338 BC. […] The settlement of 338 established a hierarchy of relationships in which the subject peoples were categorized as full citizens, citizens sine suffragio, Latins and allies. These various groups had one thing in common: the obligation to provide troops for the Roman army in time of war. The result was that the Roman commonwealth possessed enormous reserves of military manpower, and in 338 was already the strongest military power in Italy.

As it proceeded on its triumphant course, the Roman state expanded by adding an ever widening circle of dependent communities to the commonwealth. Defeated peoples were annexed with either full or partial citizenship, Latin colonies were founded, and an increasing number of states became allies. (Cornell 1995, p. 365)

Now here we have something completely different: the explanation put forward by Harris and his followers is an essentialist one – it was the Roman’s nature to be so belligerent. Cornell offers a causal mechanism instead: By turning defeated enemies into allies and, in the long run, allies into citizens of their own state, the Romans created a system that was able to expand continually, because each successful integration of a former enemy into this system increased its military resources. One might describe this as a positive feedback loop and compare it to biological modes of growth.

Independent of Cornell, Arthur Eckstein (2006) has identified the same cause for the sustained expansion of Rome, but he also contributed another perspective to the debate which seems crucial. It is important because there had been the paradox of comparative science without comparisons; by claiming that the Romans were exceptionally bellicose, one makes a statement that inevitably requires a basis for comparison, but this issue had been neglected.

Of course, it is well known how many wars had been fought in the Greek world, and the many pieces of evidence for the Greek’s appreciation of military glory are well known, too. And it is no secret that Athenian democracy had its origin in a total mobilization and militarization of the society. Nevertheless, this fact, which contributed to the aggressive stance of Athenian politics against other cities, is frequently overlooked, and it appears as if Athenian democracy arose by abstract reasoning in the lofty heights of political philosophy.

Such lines of evidence had not yet been discussed in context, from the broad perspective of a cross-cultural comparison on Rome, her Italian neighbors, and the Greek states as well as other states in the ancient world. By providing this broad comparison for the first time, by assembling a large amount of material on the role of war in the ancient world in general, rather than focusing on the Romans in isolation, Eckstein reaches conclusions which allow to see Roman militarism in a new light:

One theme, however, has come to dominate modern scholarship on this problem: that Rome was exceptionally successful within its world because Roman society and culture, and Rome’s stance toward other states, were exceptionally warlike, exceptionally aggressive, and exceptionally violent and not merely in modern terms but in ancient terms as well. […]

The present study takes a different approach. It applies to other ancient states the insights and method of analysis pioneered by Harris concerning Rome. It finds militarism, bellicosity, and diplomatic aggressiveness rife throughout the polities of the ancient Mediterranean both east and west. […] Moreover, the present study finds the origins of the harsh characteristics of state and culture now shown to be not just Roman but common to all the ancient Mediterranean great powers, all the second-rank powers, and even many minor states as well, not so much within the specific pathological development of each state (what the political scientists call “unit-attribute” theory), but rather proposes that these characteristics were caused primarily (though not solely) by the severe pressures on all states deriving from the harsh nature of the interstate world in which they were forced to exist. (Eckstein 2006, p. 3)

This gives rise to the central question:

The fundamental question is not why Roman society was militaristic and often at war, but why the Roman city-state was able to create a very large and durable territorial polity when so many other city-states failed at that task. Athens, Sparta, and Thebes all ultimately failed at it in European Greece; Carthage ultimately failed at it in North Africa; Syracuse failed at it in Sicily; Tarentum failed at it in southern Italy. (Eckstein 2006, p. 244)

And the answer put forward by Eckstein is essentially equal to Cornell’s:

It is not stern militarism but Rome’s ability to assimilate outsiders and to create a large and stable territorial hegemony that makes Rome stand out from other city-states. (Eckstein 2006, p. 245)

Rome was not alone in this liberal attitude toward outsiders. Eckstein (2006, 246f.) points out that all the Latin cities had a liberal policy in this respect, facilitating, for instance, commercial exchange and intermarriage between their citizens and allowing citizens of other cities to buy property and settle within their boundaries. The Greek poleis, on the other hand, tended toward “virulent exclusivity” and tried to restrict access to their citizenry as far as possible. For them, it would have been unthinkable to do what the Romans did, i.e., to extend their citizen rights not only to the Latins, who at least shared the common language and culture, but also to real aliens such as the Etruscans, who were not even native speakers of Latin.

Rome was particularly favored by a location that facilitated trade and economic growth, and apparently for this reason it had much better starting conditions compared to all other Latin cities, but the other crucial aspect of the rise of Rome, i.e., the ability to integrate outsiders, was common Latin heritage. So Rome could outgrow all competitors in Italy by absorbing ever more allies into her political system, but merely absorbing them would not have been sufficient. Decisive was that the system proved stable even during major crises, and this was due to a policy that maintained at least a minimum amount of consent among the allies.

The importance of this aspect can be seen by direct comparison with other powerful city-states which built alliance systems that were, in principle, comparable to the Roman one (see Baltrusch 2008 for an overview of recent scholarship on ancient alliances and empire formation; especially on early Rome pp. 9–14) but which were plagued by dissent and separatism. The maritime republic Carthage had, quite like Venice much later, a “terra ferma” in North Africa, i.e., a dominion that formed the basis for her overseas adventures. Besides the territory of the city of Carthage proper, there were a number of allies, including other cites that shared the common Phoenician origin but also the autochthonous Numidian and Libyan tribes. But the tensions between Carthage and all these neighbors were a constant theme in Carthaginian politics, and the necessity to maintain a large force to defend the homeland posed a limit on the military resources that were available for overseas operations, for instance, in Sicily. And with respect to Athens, Russell Meiggs has noted:

In the second year of the Peloponnesian war, according to Thucydides, Perikles could admit to the Athenian assembly that their empire was a tyranny. This language has shocked some modern scholars; it would not have shocked contemporaries. They knew that Athenian rule did not rest on the free consent of the allies, and I suspect that they had known this for a long time. (Meiggs 1963, p. 1)

The fact that alliance systems were vulnerable to tensions between the allies was of course common knowledge, and when Hannibal invaded Italy during the Second Punic War, part of his overall strategy was the assumption that he would be welcomed as a liberator and that the Roman alliance system would fall apart. Some communities actually defected, but the overall system remained intact, much to Hannibal’s disappointment. In this respect, he was merely repeating the experience of Pyrrhus two generations earlier, who was also faced with an essentially stable Roman alliance upon his invasion in southern Italy.

It was a principle of Roman policy to require only military service from the Italian allies. Military service was seen as honorable and it included the attractive prospect of getting a share of the spoils of war. By contrast, the Athenians initially required either military service and ships or the payment of tribute from their allies, but in the long run the demand for monetary contributions became dominant, which was a serious bone of contention. In ancient political thinking, paying tribute to another state was a sign of lost independence; this explains why this Athenian habit was thoroughly unpopular among the other members of the Delian League. The Athenians paid the price in the form of riots and, finally, dissolution of the league. And the Carthaginians frequently overstretched the patience of their allies by requiring both military service and substantial payments (Huss 2004, pp. 339–343).

Now their system enabled the Romans in a first step to create a stable hegemony in Italy. But this is not yet world domination, and the second step was the involvement of Rome in the affairs of the eastern Mediterranean from the second century BC onward. This involvement in the east finally gave rise to a unified Mediterranean world dominated by Rome, either by direct territorial incorporation as a provincia or by treaties and alliances. We cannot trace the events, spanning more than 300 years, in detail here. But it is interesting to look at the basic mechanism, because there is actually such a mechanism to be identified.

According to Eckstein, the decisive triggers were Greek calls for help, which received a positive reply from the Romans:

Our study will then conclude with an analysis of the decision by the Roman Senate and people in 200 B.C. to answer the Greek states’ calls for help against Philip V of Macedon and Antiochus III of the Seleucid empire. The two wars that followed this decision shifted the balance of power in the Mediterranean decisively in Rome’s favor and brought Roman influence and power permanently into the Greek world. In a real sense, it laid the foundation of Roman political preponderance throughout the entire Mediterranean. Yet the decision itself was of the type that we have seen throughout this study was normal (not exceptional) for a great ancient state to make when confronted by requests for help from lesser states. (Eckstein 2006, 244f.)

The network of treaties and alliances that finally led to the Second Macedonian War is too entangled to be discussed in detail, but a crucial driving force were the Aetolians, who had a long-standing conflict with Macedon and who had a major interest in getting the Romans involved.

When this happened and the Romans defeated Philip V, the Aetolians were not satisfied, however. They had hoped for a large territorial increase at the cost of Macedon, which the Romans refused to concede them. The Romans tried to establish some sort of peace order that essentially maintained the status quo before the war. The Aetolians therefore switched alliances and induced now Antiochus III to make war in mainland Greece.

So we have a basic mechanism of large networks of linked powers, linked by either long-term treaties or immediate calls for help in a situation of urgency. It was an interstate system where hostile diplomacy, armed conflict, and the switching of sides were frequent, and the growth of ever larger systems of alliances created the danger that any local conflict could easily become a major war. There is nothing specific Roman here, these are features of the ancient interstate system at large, and it could act as an amplifier of even the smallest internal conflicts within a single city.

And it is also not correct, although one encounters such views frequently, that the Romans happily took the first opportunity to impose their order, their will in the Greek east. Rather, they showed a remarkable adaptability and followed, to a large degree, the political concepts and traditions of the Greeks. Erich Gruen (1984) has shown this by detailed analyses of the arrangements made by the Romans, their treaties and alliances with Greek states, once they had become involved in eastern Mediterranean affairs.

The causes of the Persian Wars and the Peloponnesian War, as narrated by Herodotus and Thucydides, respectively, were mentioned above. Let us now look at these examples.

The Persian Wars started with the Ionian Revolt, which ultimately arose from internal dissent on the island of Naxos. The aristocratic party of Naxos appealed to Aristagoras, the ruler of Milet, for help, and he in turn asked his Persian overlords for an army. The Persians supplied this army, but Aristagoras managed to start an argument with Megabates, the commander of the Persian expedition forces. Aristagoras was in an awkward position; he switched sides and warned the people of Naxos of the imminent attack – the attack that was due to his initiative – and searched for further allies among other Greek cities along the Ionian coast. Since he felt that this was still not enough to confront the Persians in a conflict that had, by then, become a general Ionian uprising, he sought further allies on the Greek mainland. Sparta refused, which is a rare exception, but the Athenian people’s assembly were enthusiastic when they learned about the royal treasury of the Persian king, and thus the Greek mainland became involved in a major war with the Persian Empire.

The Peloponnesian War started with internal conflicts at the city of Epidamnos, and the aristocratic party appealed for help to the Illyrian natives in the area. The city thus came under pressure and asked for help at the mother city of Kerkyra, which Epidamnos had once founded. It would have been a moral obligation to provide such help, but for unknown reasons Kerkyra declined the request. So Epidamnos had to ask Corinth for help, which, in turn, was the mother city of Kerkyra. At this point, the people of Kerkyra suddenly remembered how important Epidamnos was for them, which they considered their own possession. So the intervention of Corinth in Epidamnos was seen as an insult that could only be answered by war. For this reason, they appealed to Athens for help, which was readily granted. Corinth in turn appealed to Sparta for help, and so finally the two largest powers of the Greek world – Athens with the Delian League and Sparta with the Peloponnesian League – became opponents in a war that originated in the small city of Epidamnos at the semi-barbarian fringe of the Greek world.

Seen in the light of these events, the outbreak of the Second Punic War is not the perfect illustration of Roman imperialism, in contrast to how it is usually presented. Rather, it shows just all the features that appear familiar from the examples above; in the Iberian city of Saguntum, there was internal dissent and one party appealed to Rome, making her the arbitrator and protector of the city. In addition, Saguntum had conflicts with surrounding tribes who were allied to Hannibal. When open war between Saguntum and the tribe of the Turdetani broke out, the latter appealed to Hannibal, while the Saguntines sent envoys to Rome. That Saguntum was located south of the river Ebro and thus in a region that had been defined as Carthaginian, rather than Roman, zone of influence did not help to deescalate the situation either, and now the whole system was ready for a major war.

But why then did the large states agree to be drawn into the messy affairs of minor powers? They knew that far-ranging and destructive wars could ensue, and they frequently also knew that the legal or moral justification for their intervention was weak, as in the case of Athenian help for Kerkyra as well as of Roman help for Saguntum. But the irresistible benefits from the perspective of the large powers were always the same:

  • Reputation – it conferred prestige to be a widely accepted helper and arbitrator, and international prestige is a value in itself (for ancient as well as for modern governments).

  • Increased radius of operations – these calls for help were an optimum pretext for promoting one’s own interests in distant regions, spoils of war and other material benefits included.

  • Competition – if you decline this request for an alliance, someone else will accept it and reap the benefits in your place.

So in these respects, Rome is typical rather than exceptional. But it was the rise of Rome to the single dominant power in the Mediterranean that effectively ended this violent interstate system with its frequent wars and destabilizing alliances.

Without doubt, Roman domination had its own adverse effects. In particular, outside of Italy the Romans did not continue their system of alliances without tribute payments. Rather, they imposed taxation on their provinces abroad, and the tax burden was probably the single most important cause for the riots and separatist movements that occurred in some instances in the Roman Empire. But in the overall balance, the empire remained remarkably stable, and the benefits of the Pax Romana seem to have been real, rather than being perceived as mere propaganda. The reasons for the empire’s stability are manifold, but one aspect stands out, by direct comparison with the violent interstate system that had prevailed previously: the removal of the background of constant warfare brought in itself substantial economic benefits and enabled a period of general prosperity. To conclude this section, I will present three illustrative examples from the Roman east.

In 167 BC the federation of Lycian cities (in the southeastern corner of modern Turkey) had been declared “free” by the Romans – meaning that Rome became the protector of their independence from the former overlord Rhodes, after a series of Lycian revolts. The result was an enormous building activity, the traces of which can still be seen today (Marek 2010, 291f.). Around the same time (ca. 170 BC), the Pergamon Altar was built, one of the most impressive extant monuments of antiquity. Again chronology reveals the causal nexus: after his defeat by the Romans and the peace treaty of Apameia in 188 BC, Antiochus III had to withdraw from Asia Minor. Thus, the kingdom of Pergamon, not yet a Roman province but under protection of Rome, could recover from the previous wars, and an ambitious building program was started at the capital. A strikingly similar pattern is observed one century later, when the Near East had also come under Roman influence:

What is now certain, however, is that Petra, as a city with monumental architecture and rock-cut facades, belongs in that period in the history of the Near East when Roman domination was assured, but Roman direct rule was either absent or still relatively lightly imposed. The royal monuments of Commagene belong in this period, if in the earliest phase of it; but also do the temple of Bel at Palmyra, Herod’s Temple in Jerusalem (as well as his other major monuments), the temple of Baalshamin at Sia’ and, as will be seen, the temple of Zeus at Gerasa. The major constructions of this period were sometimes royal creations, as in Commagene or Judaea, but others were expressions of the culture of local communities, as in Palmyra, Sia’, or Gerasa. (Millar 1993, 407f.)

4 How to Run an Empire?

It should have become clear that a single master plan for world conquest is not the explanation for the emergence of the Roman Empire. The power and influence of Rome grew in a piecemeal fashion, as the sum of many individual episodes. And the patchy nature of the Roman possessions was also continued in the imperial era, when territories with quite different legal status became collectively called the Imperium Romanum. Even the question what the empire really was defies a clear modern definition: it was not simply the personal possession of a king, like the Hellenistic monarchies. It was also not simply the territory of the Roman people because the latter, i.e., the ager Romanus resp. the legally defined Roman homeland in Italy, was never expanded beyond central Italy. The best approximation of a legal definition in modern terms describes the empire as an alliance of cities with Rome as the senior partner. So the administrative structure reflects the mode of growth of the empire, and it had three levels: the cities with considerable local autonomy, the provinces, and finally the emperor. The latter two represent the “imperial” or “Roman” administration of the empire, but it is uncertain what this really means. Our sources do not explicitly explain the workings of the imperial administration; we are lacking texts that would provide an organization chart or handbook of administrative procedures. But what we do know is that the bureaucratic apparatus was small by modern standards. There were no large bureaucracies, neither in Rome nor in the provinces. The emperor ruled essentially with the help of a limited number of friends, advisers, and secretaries. And the same system of minimal government was repeated by the individual governors in the provinces, who also had only a small staff. Some scholars even deny that the modern notion of an administration applies at all to the Imperium Romanum (especially on Roman Asia Minor, see Marek 2010, 453f.; for a general discussion of the emperor’s role, Millar 1992 is essential).

The absence of a professional bureaucracy indicates that the empire cannot be understood in terms of a modern state. A further example which provides evidence for this is the issue of Roman strategy. The problem is that there were no general headquarters of the army, no ministry of defense or state department, no secret services, no permanent embassies or professional diplomats, no institutes for political science or international relations, and no think tanks or military academies. The whole institutional framework that is essential in a modern state in order to formulate the strategic aims is lacking.

Nevertheless, Edward Luttwak published his “Grand Strategy of the Roman Empire” in 1976, and the debate about Roman strategy is still influenced by this work (see Campbell 2010; Heather 2010 for recent discussions).

Luttwak has analyzed the military arrangements of the Roman Empire, in particular with respect to the borders, and he considers three distinct phases of Roman strategy:

  • Phase 1 in the time of Julio-Claudian emperors (27 BC–68 AD): The system is based on client states and mobile armies, a broad buffer zone provides security, and the mobile army is concentrated at several points (Luttwak 1976, pp. 7–50).

  • Phase 2 from the Flavians till the Severans (69 AD–235 AD): A system of “scientific frontiers,” i.e., short frontiers selected for optimum defensive qualities, and preclusive defense. The empire becomes a sort of fortress with precisely defined perimeter and limes fortifications (Luttwak 1976, pp. 51–126).

  • Phase 3 in the later third century until Diocletian (235 AD–305 AD): Fixed frontiers are abandoned and replaced by defense in depth; the mobile armies are located in the interior and serve as a large strategic reserve that can operate wherever needed (Luttwak 1976, pp. 127–190).

Some of the facts are uncontroversial. For instance, a number of semi-independent kingdoms were successively transformed into regular provinces during the course of the first century. However, the majority of scholars have denied that one can identify clear and elaborated systems of Roman strategic overall planning.

Any discussion on the Grand Strategy of the Romans must start with the basic evidence: the number of troops, the borders of the empire, and the distribution of the troops within that territory. But the problems start here already: We do not know, for instance, which factors limited the size of the Roman army, since ancient sources do not discuss these matters. In general, it is assumed that financial, rather than demographic, reasons set the limit for a comparatively small army of about 300,000 men under Augustus and his successors for the next 250 years or so. But this is just conjecture, and there is another possible explanation; a large army posed a potential political threat. In the final decades of the Republic, soldiers had repeatedly played an active role in military and political matters, and generals, including Octavian, had been forced by mutinies to follow the wishes of their soldiers (Keaveney 2007; Kienast 2009, 320ff.). Whether these experiences played a part when Augustus finally designed the army of the principate is unknown but conceivable at least. What is known is that in any case the number of troops garrisoned in one place, under the command of single provincial governor, was restricted. Large troop concentrations were frequently the crystallization points of attempts to usurp the emperor’s purple by ambitious generals. In short, domestic policy may have been at least as important as the strategic response to external threats for decisions on troop distribution.

The notion of a coherent Grand Strategy of the Romans suffers also from the problem that the required institutions were not there, as discussed already. Implementing a strategic doctrine in the modern sense would have required the collection, colligation, and evaluation of large amounts of information – it is unclear who should have done this.

Another problem is the issue of maps. Looking at the Roman Empire from a bird’s eye perspective, as we can do using modern maps or satellite pictures, certain features appear rational, such as the use of deserts, mountain ranges, or large rivers for defining the borders. But it is unknown whether this way of looking at things was available to the Romans, since it is contentious whether they used maps comparable to modern ones. This problem obviously affects the question why they made particular border arrangements.

The Roman agrimensores made exact measurements of the terrain on a small scale. These techniques were used, for instance, for building roads, for laying down the city plan of a newly founded colonia, or for constructing a military camp with its regular structure. These techniques were obviously also used to produce the giant map of the city of Rome from the time of Severus (ca. 200 AD). It is striking how this graphic representation appears familiar to a modern observer, almost two millennia later (Koller et~al. 2005). So when the Romans were able to do such things for public display, would they not have used the same techniques for strategic planning?

But getting the topography of a city right is one thing, getting the topography of a whole continent right quite another. Under Augustus, there was also the world map of Agrippa on display in Rome – this may have been what we are looking for, but some scholars have denied that this was a real map and suggested that it was an itinerarium (Kienast 2009, p. 264, fn. 187), i.e., a list of places and the distances between them. Such itineraries served to break down the two-dimensional topography of an area into linear, one-dimensional relations between places.

Since the sources do not tell explicitly if and how maps were used for strategic purposes, other approaches are needed. Christian Hänger (2001) has analyzed ancient sources with respect to the geographical knowledge of the acting persons, insofar as it can be inferred indirectly from the text. When Tacitus reports, for instance, on the campaigns of Tiberius, Drusus, or Germanicus in Germany, his own account of the landscape is fragmentary and superficial. But, as Hänger has shown, the Roman generals’ geographical knowledge must have been much better than the scanty fragments that made their way into the historian’s account. The choice of optimum marching routes and the clever use of rivers as supply lines betray familiarity with the area of operations (Hänger 2001, 180ff.). According to Hänger, the Roman commanders must have possessed at least very precise mental maps. This is not yet proving the existence of physical maps, but it makes them plausible in any case (see also Sheldon 2005, pp. 148–150 on Roman strategy and maps. According to a recent analysis by Fedi et~al. 2010, the so-called Artemidorus Papyrus is probably authentic; it apparently contains a map of the Iberian peninsula from the first century AD).

Thus, it seems possible to cure one important moot point of Luttwak’s theory, but this does not end the discussion. One of the most severe criticisms came from Benjamin Isaac (1990), who has offered an alternative explanation for the military arrangements of the Roman Empire. Isaac rejects a Grand Strategy and a defensive organization of Roman troops altogether. But his major point is not the question of a feasibility of Roman overall strategy in a technical sense – the issue of maps and so on – rather, he sees a completely different motivation at work; according to him, the Roman army served mainly aggressive purposes to enable the emperors to celebrate themselves as triumphant conquerors and also as an instrument of extortion and oppression of the subjected peoples. The influence of Harris’ views is obvious.

Luttwak’s perspective was biased because he tried to transfer concepts from modern strategic studies to the ancients. But Isaac’s perspective appears likewise biased, because his conclusions on the particular area, i.e., the Roman Near East, cannot provide a model for the interpretation of the empire as a whole, with borders in vastly differing regions of the world, facing a multitude of different local conditions and challenges. The Near East was the place of the Jewish War and other Jewish revolts, but these events are the exception rather than the norm. The revolt of Arminius, for instance, that gave rise to the famous battle of the Teutoburg Forest was not some kind of Pan-Germanic and anti-colonial liberation movement. Rather, it was a military coup of an ambitious leader who calculated that being a Roman officer was nice, but being a Germanic king even nicer. In general, the Roman Empire is characterized by the rarity of riots that are “nationalist” in modern parlance.

The allocation of the legions shows impressively that large interior areas of the empire were virtually military free, in particular the peninsular areas of modern Spain, Greece, or Turkey. The army appeared to be concentrated in two particular border regions, the northern one along Rhine and Danube and the eastern one. This allocation cannot be reconciled with suppression of internal riots being the main purpose of the army.

Leaving aside Isaac’s extreme view, his denial of any defensive considerations on the part of the Romans, a consensus exists among the critics of Luttwak’s theory. The attempt to transfer the modern concept of a coherent, systematic, and long-term strategic overall planning is generally regarded as failed.

That Roman arrangements may have been piecemeal rather than systematic can be seen, for instance, by comparing the limes systems of the neighboring provinces Germania Superior and Raetia. Both are located in the same type of Central European landscape, and both have faced essentially the same type of enemies. Nevertheless, the Germanic limes consisted in its final stage of a wooden palisade, a ditch, a rampart, a connecting road, and, finally, on the inner side of the whole arrangement, a line of watchtowers. By contrast, the adjacent Raetian limes was formed by a stone wall into which the towers were integrated, while the connecting road was behind them. No plausible explanation for these differences in terms of different tactical necessities exists, and it seems that there were simply different regional traditions at work (discussion of various limes systems from the perspective of intelligence and communication in Sheldon 2005, pp. 199–249).

In philosophical terms, one might describe Luttwak’s approach as a logical reconstruction. He took apart pieces from Roman military history and reassembled them to fit a modern strategy analysis. But since he used modern concepts and notions, his work yields no genuine explanations for the Romans’ decisions because their original perspective is no longer considered in his reconstruction:

When Edward Luttwak […] analyzed the defense policy of the Roman Empire using the vocabulary of modern military structures, he produced an interesting conceptualization for readers at the Pentagon, but I am still not convinced he brought us any closer to an understanding of the Roman mentality. Roman activities were often messy, unprofessional, and even unsuccessful, but we do less damage to the historical record if we leave them that way. Trying to incorporate them into a grand strategy that may not even have existed may be more satisfying to us intellectually, but it is ultimately less accurate. (Sheldon 2005, p. xvi)

So rather than assuming long-term and “scientific” planning, we should see Roman decisions as frequently being ad hoc, opportunistic, and based on the personal idiosyncrasies of individual emperors. It was, in particular, Fergus Millar (1982) who stressed the highly personalized form of Roman government, the lack of large and inert institutions, where the individual preferences of emperors necessarily became a major determinant of Roman politics and warfare. And these preferences were not always of a “rational” or “objective” nature. Rather, ideology and tradition, thinking in terms of glory and precedent, played a great role. The Roman emperors’ desire to emulate Alexander the Great defies the attempt to analyze it using the tools of modern strategic studies, but exactly this desire was one major reason for the repeated wars between Rome and her eastern neighbors.

This constant worship of the classical tradition, of the glorious examples of the past, does not only affect Roman texts with respect to their usefulness for us. It may have likewise affected the contemporaries’ ability to comprehend their own time. Herwig Wolfram has identified an “incorrect theory” with respect to Germanic peoples as one of the major causes for the difficulties the Romans experienced when the Germanic world set itself in motion in late antiquity. When thousands of Lombards crossed the Danube under Marcus Aurelius in 167 AD, the Romans knew – at least since the beginning of the Christian era – that their original places of settlement were at the lower Elbe River, hundreds of kilometers to the north:

But nobody considered to draw any conclusions from this, such as asking, for instance, whether movements within Germany might have been the cause for the outbreak of the terrible fights, and if perhaps even worse things were to come. […]

Such an approach would have required an understanding of the barbarian world that was not available, rather, one liked to believe that there were no and could be no new barbarians. […] How should a government in Rome have taken the correct preemptive measures, when its experts on foreign matters were literates, who designated Goths, Vandals and Huns alike as ‘Scythians’, equaling them to that people of the southern Russian steppe which in reality had become extinct long ago. For the situation at the Rhine, one similarly used traditional categories: Alemanni and Franks continued to be seen as Germans of the type encountered in the early imperial era or even – in accordance with the pre-Caesarian custom of the Greeks – as ‘Celts’ or ‘Celtoscythians’. (Wolfram 1990, 70f., transl. SB)

How far this literary and antiquarian approach actually affected Roman decision making is a matter of debate. Susan Mattern (1999, pp. 1–2) pointed out that the literates, who produced such works, were frequently amici of the emperor, i.e., persons very close to the center of power. Thus, she concluded that the level of strategic thinking found (or rather not found) in their works reflects the debates among the emperor and his companions. However, as one reviewer of her book has argued (see Sidebottom 2003), exactly because the tendency of literary stylization in such texts is known, we cannot use them to make direct inferences about the debates that were actually led.

In any case, Kimberly Kagan has warned that one should not overstretch the criticism of Luttwak’s book and she pointed out that when the definition of Grand Strategy becomes too narrow, it finally becomes useless even for discussing the behavior of modern states. In particular, she noted that the crucial aspect is not how far the Romans conformed to modern definitions of strategy. Rather, one should look at what they actually did:

The patterns of troop movements also show clearly that imperial decision-making about grand strategic issues occurred even without visible long-term planning. Emperors restored distributions even after intervening events such as wars, rebellions, and imperial successions had disturbed them. […] Emperors worried about the stability or security of provinces when they conducted major operations. The successive replacement of legions moving off to war shows that emperors thought about how their activities on one frontier (in crisis or for conquest) affected the whole empire. Emperors made decisions about how to allocate resources to meet objectives empire-wide, and thus definitely thought about grand-strategic issues. The grand strategy of the Roman Empire can be studied as long as we ask questions that the available sources support. (Kagan 2006, p. 362)

Besides troop movements, the issue of bridges is a further example: For long stretches of time, there were no bridges across the two major border rivers in Europe, the Rhine and the Danube. The Roman technical expertise to build such bridges is out of question, and the deliberate refusal to build them is even more astonishing given their potential benefits; they would not only have facilitated commercial exchange in itself, they would also have facilitated the collection of import and export taxes by channeling this traffic. Nevertheless, the function of large rivers as an obstacle for barbarian intruders was obviously ascribed a higher value than other criteria.

In conclusion, it is clear, on one hand, that the Romans were not completely blind to reality; they included issues of defense and security into their plans. On the other hand, the concept that the Romans followed coherent and rational strategic systems with long-term and centralized planning is rejected. In particular, the idea that there was a succession of three distinct, clearly defined military doctrines is generally regarded as the weakest point of Luttwak’s analysis. The shape of the empire, as it emerged during its history, may appear rational to us, with respect to the choice of river borders and so on, but it is, after all, a product of chance, the accumulation of numerous decisions which were made for a whole range of different reasons. What we have here, in effect, is a beautiful analogy to results of biological evolution. It may look like a product of rational design, but it is a product of chance nevertheless.

5 Concluding Remarks

Two central questions in Roman history appear quite similar from a formal perspective:

  1. 1.

    How can we explain the rise of Rome, which policies and strategies were used by the Romans when they conquered their empire?

  2. 2.

    How can we explain the shape and structure of this empire, which policies and strategies were used by the Romans to defend and/or enlarge it?

Despite the formal similarity of both questions, there are deep differences with respect to the tasks that modern historians have to confront. In the case of the rise of Rome, there is abundant ancient evidence on the detailed course of events. One may even say that there is an overabundance of evidence because so many individual events are known that somehow may have contributed to this process. Thus, when a historian wants to explain the rise of Rome, a major part of the task is to select the pieces which are crucial, to form a coherent narrative from the available material by declaring some points as relevant, but others as irrelevant.

In contrast, the issue of Roman strategy once the empire had emerged is quite another story because no ancient sources allow for a direct insight into a possible overall strategy of the Romans. Therefore, in this case the historian’s task is to assemble as many clues as possible from the fragmentary literary and material evidence. The various problems concerning ancient sources, as discussed above, are therefore more pertinent here.

One tool that plays a role in both fields is the comparative approach. This should be stressed, because comparative approaches, like the related issues of world history or universal history, are still a kind of fringe activity in modern historiography, which is still organized along traditional boundaries of eras and regions. To explain the rise of Rome, the direct comparison with other ancient states has destroyed the myth of a particular Roman attitude toward war. In addition, comparisons with the modern world occur in both fields of inquiry. For the issue of the rise of Rome, the application of the notion of an interstate system, borrowed from modern international studies, has proved fertile. In contrast, the transfer of the concept of a Grand Strategy to the Roman Empire, borrowed from modern strategic studies, was not successful in the eyes of a majority of scholars of the ancient world.

But even such failure can have its merits. The comparative method can either demonstrate some underlying common principles or serve to highlight the differences between apparently similar trajectories and phenomena of history.

To illustrate this point and to conclude this chapter, we will have a look at ancient China. A comprehensive general theory of ancient empires does not exist yet, since such broad perspectives are rather unusual and unpopular in historiography (Pomper 2005 is focusing on modern empires and does not even mention the Imperium Romanum; Münkler 2005 has a few observations pertinent to the theme). Nevertheless, a number of scholars are working in this direction; Walter Scheidel, who is one of the pioneers in the field and has recently edited a whole volume dedicated to the detailed comparison of Rome and China (Scheidel 2009b, see also Morris and Scheidel 2009), notes:

Two thousand years ago, perhaps half of the entire human species had come under the control of just two powers, the Roman and Han empires, at opposite ends of Eurasia. Both entities were broadly similar in terms of size. Both of them were run by god-like emperors residing in the largest cities the world had seen so far, were made up of some 1,500 to 2,000 administrative districts, and, at least at times, employed hundreds of thousands of soldiers. Both states laid claim to ruling the whole world, orbis terrarum and tianxia, while both encountered similar competition for surplus between central government and local elites and similar pressures generated by secondary state formation beyond their frontiers and subsequent ‘barbarian’ infiltration. (Scheidel 2009a, p. 11)

The point at the end of this quotation is a very interesting one: the process of secondary state formation in loosely integrated tribal societies, which arises from the contact of a large, centralized empire with its barbarian neighbors, is one of the most important recurrent features in the history of the premodern world. There is no doubt that both Roman and Chinese studies could benefit from a cross-cultural perspective that, in the case of Rome, may contribute to a causal explanation of the late-antiquity phenomena known as “Völkerwanderung” in German or “Dark Ages” in English.

But returning to the issue of formation of the Roman Empire, a comparison with the first unified Chinese Empire is also illuminating. Both empires arose within a violent interstate system of competitive states engaged in unstable alliances and constant warfare among each other. The Chinese era between the fifth and the third century BC is aptly called “Warring States,” and incidentally, it covers a span about as long as the formative time between the First Punic War and the start of imperial government under Augustus. The state of Qin finally emerged as a superpower with a military potential that had no equals, and its ruler was from 221 BC onward the sole ruler of the Chinese world.

Despite these similarities, the means for maximizing the military potential were quite different in both cases. As we have seen, Rome grew in an extensive mode, by adding communities to the political system. There was no forced homogenization, communities retained a substantial degree of internal autonomy, and accordingly, the central administration was small during both republic and empire. It was small compared to a modern state, but also compared to the first Chinese Empire, because Qin became the most powerful state in its world by intensive, rather than extensive, growth:

Central to the Qin reforms was the grouping of the population into units of five households that were each responsible not only for providing the squads of five recruits that formed the building blocks of Qin armies but also for mutual surveillance. […] To ensure that the maximum amount of land was brought under cultivation, Qin also penalized households with adult sons living at home. These penalties forced sons to establish independent households and to cultivate their own allotments of land in order to support them. In tandem with this step, Qin also divided its territory into a grid of blocks, each of which was sufficient to support a family from the food produced on it. This reshaping of the countryside in order to ensure the maximum extraction of the resources for war was given physical expression through a system of paths forming a rectangular grid over the crop lands of the state. Finally, the government financed its war making through a head-tax imposed on the population.

Qin carried out this vast effort at social and economic engineering through the creation of an equally extensive administrative apparatus. […] To control this system, Qin established a bureaucracy capable of extending the central government’s reach down to the local level. (Rosenstein 2009, 25f.)

So despite similarities in the ways to empire, it seems that some deep-rooted differences between Western and Chinese culture actually may be accounted for by explanations that date back 2,000 years. While in Western political discourse the ideal of plural, independent entities plays a central role (see, e.g., the hostile reactions in all European countries against the concept of further integration and unification), Chinese politics – imperial or communist – was always much more focused on the ideal of a strong and heavily centralized state. And it is certainly no coincidence that Fernand Braudel, who coined the expression longue durée for such long-term continuities, chose the Mediterranean world as a prime example in his studies.