IV

IMPERIAL ECOLOGY

Next comes the earth, that one part of nature that for her many gifts to us we honour with the name of Mother. She is our realm, as the sky belongs to the gods. She welcomes us when we are born, nurtures us as we grow, and when we are adults sustains us always.

(Pliny, Natural History 2.154)

From its foundation to the Arab conquests the story of Rome was played out over a millennium and a half. At first expansion was so slow that few outside Italy can have noticed it. But by the reign of Augustus the empire was bounded by the Atlantic to the west and the Sahara to the south, its northern frontier bisected temperate Europe, and its eastern edge was extended deep into western Asia. There the frontiers more or less rested until disintegration began at the end of the fourth century, once again slow at first but eventually collapsing into the Aegean world of seventh-century Byzantium. That fifty-generation tale of rise and fall is an epic one in human terms.

Geologically, however, a millennium and a half is the blink of an eye. The Roman Empire was a bubble that grew on the surface of the pond and then burst. During this time the physical environment of the Roman world—its landforms and climate in particular—hardly changed. New crops and methods of agriculture spread, but they had only a little impact on the landscapes Rome ruled over. Had Romulus been transported seven centuries forward at his death (rather than taken up into heaven) he might well have been amazed at what his heirs had achieved, but he would not have been puzzled at how they did it. All this is hard to imagine today, living as we do at the end of two centuries of accelerating technological change, change that is having major impacts on the entire biosphere and moves now at a pace hard to adjust to psychologically even in our own brief lifetimes. This chapter explores the long-term stability of the ancient world and the slow secular changes against which the whole of Rome’s imperial story was played out.

The Environment in Classical Antiquity

Let us begin with the visible. The coastline of the Mediterranean 3,000 years ago was hardly different from what it is today. Slight changes can be spotted around the mouths of the larger rivers: the harbour of Ephesus is now a few miles from the coast, and half of Ostia, the port of Rome, has been washed away. Just off Pozzuoli in the Bay of Naples, a series of luxurious villas lie just a few metres under water. So does the great harbour of Alexandria. But these are marginal changes in highly susceptible locations. Sea levels did begin to rise gradually at the end of antiquity, but the only regions where this had a marked impact were low-lying areas such as the Fenland of eastern England and the areas around the Rhine mouth in the Netherlands, where late antique villages are built on low mounds, terpen, for protection against floods. Unsurprisingly, the ancients had no sense of geological time or incremental environmental change. There was almost no ancient science of seismology, and the explanations suggested for earthquakes were underground counterparts of those developed for meteorology.¹ A few writers were so committed to this steady-state idea of the world that they believed marble would, eventually, grow back from where it had been quarried.² Their world was eternal; the gods had wandered in the same forests and mountains they knew.

The Mediterranean is, in fact, shrinking, as the African tectonic plate moves northward. But this is happening very slowly. Tectonic movement generates vulcanism in Sicily, the Lipari Islands, and Campania, and earthquakes in central Italy, central and southern Greece, and western Turkey. There are extinct volcanoes in other parts of the Roman world—around Rome, for example, in central France, and southern Scotland—but the ancients had no memory of their eruptions. Volcanoes and earthquakes occurred in antiquity more or less where they occur most often today. Etna, into which the philosopher Empedokles reputedly threw himself, Vesuvius, which buried Pompeii and Herculaneum, and Santorini, which did the same for the Bronze Age city of Akrotiri, remain active today, indeed the most spectacular volcanoes of the Mediterranean world. Major earthquakes have struck in recent memory at the Isthmus of Corinth, where Poseidon the Earthshaker had his greatest sanctuary; in Aegean Turkey, the cities of which received five years’ remission of tribute in AD 17 from the Emperor Tiberius to help them rebuild themselves after a devastating quake; and in central Italy, where a series of quakes in the middle of the first century AD perhaps inspired Seneca to write our first surviving discussion of them, in the sixth book of his Natural Questions.

Climate change moves at a faster pace. But in climatic terms, too, the ancient Mediterranean was very similar to the one we know today.³ We live within the same interglacial period as the Romans, the Holocene, which began around 12,000 years ago with the retreat of the European glaciers and the northward expansion of the Sahara Desert. As the region warmed there were consequent movements of plant and animal species. Such movements are slow: in botanical terms the Mediterranean can be considered as still in postglacial recovery, and not all its native species of plants are yet well adapted to the current climate. Around 6,000 years ago the Mediterranean basin became significantly warmer, establishing today’s pattern of mild wet winters and warm dry summers. ‘Mild’ means no long freezes which kill many species of tree and plant. ‘Dry’ means the Mediterranean as a whole was, and remains, an arid environment. There was never sufficient rainfall to support either dense forests or the grasslands on which herds of ruminants such as cattle, horses, and bison depend. Some parts of the Mediterranean world are exposed to droughts severe enough to cause many crops, including wheat, to fail as often as one year in four. Droughts of that kind are not predictable, and have knock-on effects on species that depend on susceptible crops. Humans that farm are among such species. Classical civilization was built in the shadow of scarcity and risk.⁴

The Roman Empire originated in the Mediterranean basin. But from the end of the last century BC, it had expanded into adjacent ecological zones. The climate changed most dramatically as one went north or south. This ecological gradient had economic consequences, since many of the central components of the polite culture adopted by local elite members across the empire remained Mediterranean in character. Wine was the alcoholic drink of choice, even where it was easier to produce beer: in the early first century AD some Mediterranean producers grew rich producing wine for export until grape varieties were developed that could survive in the Rhineland and even southern Britain. Olives could not (and still cannot) be cultivated in regions susceptible to frost. Yet olive oil was essential not only for cooking but also as fuel for lamps and for use in Roman bathing, where it was rubbed onto the skin and then scraped off along with any dirt. Olive oil was consequently traded northwards in great quantities.⁵ The southern half of the Mediterranean is notably warmer. From the late second century BC, great quantities of grain were being exported from modern day Tunisia, Sicily, and Egypt to cities in the northern half of the Mediterranean.⁶ By the early first century AD olive oil production had also increased in southern Spain and various parts of North Africa.⁷ North of the Mediterranean basin, temperate Europe had harsher winters and much more plentiful rainfall. That made it a much better area for raising large domesticates. Rome’s European provinces would come to supply much of her cavalry. The accumulating evidence of animal bones found on Roman period sites also shows much more beef eaten north of the Alps, with sheep and goat most evident in the assemblages from sites in the more arid south and east of the empire.⁸ Studies of faunal material (bones), of seeds and other biofacts, and of container amphorae all also show how the most powerful and privileged members of Roman society—local aristocrats and soldiers for the main part—were able to consume more or less what they liked wherever they were. The main limit on exchange across these sharp ecological contrasts (ecotones) was the cost of transport. Even the journey to north Italy made olive oil so expensive, according to St Augustine who was born and brought up in North Africa, that it was too expensive to burn lights all night long. Traffic beyond the Mediterranean basin was blocked at several points by mountain ranges. A few cities, located at the southern terminal of north–south river valleys or mountain passes, grew rich on trade: Aquileia and Aosta in Italy, Arles and Narbonne in southern France still impress visitors with their Roman period monuments.

The climate of Holocene Europe has not been completely stable. A relatively warm period in the Middle Ages was followed by the Little Ice Age which ran from 1300 to 1800 and was at its coldest at the end of the seventeenth century. Mean temperatures were perhaps a degree or more below those today, but this was enough to make the Thames freeze over on a regular basis. Evidence is mounting for a Roman Warm Period, one that perhaps raised the mean temperature as much as two degrees above those of today.⁹The proposed peak is around 150 AD, with temperatures dipping until they began to rise again at the start of the Medieval Warm Period, perhaps around 900 AD. Geophysical evidence from ocean sediments and ice-cores, and tree-ring data, is supported by literary and archaeobotanical indications that some plant species existed further north or at higher altitudes during the early Roman Empire than they do today. The reality of this phenomenon remains very controversial. Unlike the early modern cold period, any change in antiquity was too slow to be noticed by ancient observers. But it may have been important. It has been pointed out that this Warm Period would coincide chronologically with the empire’s furthest northward extent and with the Roman urban maximum. Might a period of warming have increased the productivity of southern Mediterranean agriculture, and made it easier to adapt crops for northern Europe at just the right moment? And might the subsequent cooling have put pressure on Roman agriculture (weakening the empire)? Or else on the barbarian peoples living north of the empire (driving them south)? Investigating these apparent correlations is a priority for future research.

A World of Farmers

Most ancient writers took their physical environment as a given. But they were well aware of the transformative power of one human activity, and that was farming.

Agriculture was an invention of the Holocene, one made independently on at least half a dozen occasions, around the globe. Each invention was based on different combinations of crops—cultigens—that could together supply the carbohydrate needs of humans, and some of their protein. Societies that experienced a Neolithic Revolution were utterly different from those that had preceded them. Population levels rose, permanent settlements were almost always necessary, and in these growing villages and cities, new social discipline was required. As Neolithic societies achieved a new order, so too did Neolithic landscapes. Areas suitable for agriculture were cleared, restricting hunting to marginal territories. Meat eating began to decline. Populations who lived on high-carbohydrate diets, in closer than ever proximity to each other, were less healthy than their ancestors. The domestication of successive animals improved the protein supply, but brought more diseases. Our leap down the food chain came at a heavy price, but once population levels had risen it was in effect irreversible.¹⁰

The nearest farming revolution to the Mediterranean world was also the earliest on the planet. It began around 7000 BC in the Near East in what is sometimes called the Fertile Crescent. This broad band of territory arcs from Jordan up through Syria and down through modern Iraq to the Persian Gulf, skirting the northern desert of the Arabian Peninsula. Successive innovations had their origins in this region, and in its surrounding upland margins, especially in Anatolia. The move from gathering to cultivating wild crops was followed by the domestication of animals first for food and later for hides, wool, milk, and traction. Arable farming moved across Europe at the rate of about 25 kilometres (15 miles) a generation, reaching the Atlantic by 3000 BC. The first farmers used tools of flint or obsidian, with handles of wood and bone, and they had no traction animals. Forests were hard to clear with stone axes, and wooden scratch ploughs (ards) were most effective on lighter soils. The main crops farmed were husked varieties of wheat, emmer, and einkorn, and in the most arid regions barley. Grains were supplemented in their diets by pulses and some green vegetables. Hunting and fishing made up a tiny proportion of most people’s nutrition. Agriculture spread more rapidly in the Mediterranean world than in temperate Europe for two reasons. First, communications were easier across the islands and coastal settlements of the inland sea than through the forests and mountains of Europe. Second, the aridity of the Fertile Crescent meant the ecological distance was less to the dry Mediterranean than to colder and wetter regions. These two factors together explain why population growth, cities, and states came to Mediterranean Europe before they reached the continental interior. But the Mediterranean only had a head start. Once northern populations had mastered the techniques of farming and developed ways to unlock the much greater potential of Europe’s deeper soils, more abundant, and more dependable rainfall, the Mediterranean would lose its advantage. Today southern Europe is the poorer half of the continent, the recipient of subsidies provided by richer economies of the north. That shift took place in the Middle Ages. Classical antiquity is very largely the history of the period in which the Mediterranean kept ahead of temperate Europe in this respect.

The great story of the Mediterranean Holocene is one of successive movements from east to west. New cultigens, domesticated animals, technologies, and social forms all had their origins somewhere or other east of the Mediterranean world, most around the Black Sea, in Anatolia, Egypt, and Mesopotamia. Greek and Roman writers treated the Mediterranean as the normal centre of the world. The further one went away from it, the stranger were the peoples and plants and animals one encountered. But in reality the Romans’ orbis terrarum was just one of several peripheries to the continental mass of Eurasia and Africa.¹¹ The Old World, in ecological and civilizational terms, has always had a common history, within which Mediterranean and European history has always been a secondary development.

The reasons are fundamental. Plant and animal species move most easily within the same latitudes, where mean temperatures are broadly similar. Migrations of species—both the re-colonization of Europe after the glaciers retreated, and the spread of the cultigens and domesticated animal species created by the first farmers—took place most easily between similar environments. The expansion of the Sahara accentuated this effect, creating a barrier between the Mediterranean and the rest of Africa. Ecologically, the Mediterranean world is a long corridor leading westwards out of western Asia. The climate becomes significantly wetter further west owing to the proximity of the Atlantic. Within the Mediterranean this is expressed in the differences in rainfall between the wetter west-facing coasts of Italy, Greece, and Turkey, and their more arid east-facing coasts. The further west along the corridor a species moved, the greater the contrast with the ecology within which it had originated. It is as if the corridor was at a gradient, sloping upwards, so making westward progress increasingly difficult.

Domesticated animals first appeared in Europe in the third millennium BC and again spread westwards. The origins of domestication were again mostly in the Near East, and the main domesticates were widespread by the start of the last millennium BC. Oxen and horses provided traction. Cattle, sheep, goats, and pigs provided meat. Cattle, sheep, and goats might be milked. Sheep and goats provided wool. All might provide bone and leather, and in sophisticated agricultural regimes manure as well. By the end of the second millennium BC, humans were using one or another variation of this complex of cultigens and domesticates across Europe and the Mediterranean. Geese had been domesticated in ancient Egypt. Chickens, descended from jungle fowl in the Far East, appeared sometime in the middle of the last millennium. (There are no chickens in the Iliad, but Socrates’ last words were that he owed a cockerel to the god Asclepius.) Camels, domesticated much earlier in the Arabian peninsula, also moved progressively further west during the last millennium BC. Rabbits were confined to the Iberian peninsula until the turn of the millennium. Small animals might seem insignificant, but they were valuable as they might be bred fast and fed cheaply. In a world without refrigerators, small animals also posed fewer meat storage problems.

The main technological innovations in agriculture also preceded Rome. Most important was metallurgy, also invented in the east. Gold was easiest to extract, but was nearly useless for making tools. Copper and then bronze working began in the Near East in the middle of the fourth millennium BC. What really made the difference to agriculture was the appearance of abundant iron tools, cheaper to produce and harder wearing than other metals. Ironworking was almost certainly discovered in north-west Anatolia towards the end of the second millennium BC: from there it spread throughout the Old World, reaching the Yangtzee, southern India, and Scandinavia by the middle of the last millennium BC. Iron agricultural tools were especially important in northern Europe where they made forest clearance and the cultivation of heavy soils much easier. The growth in the availability of iron tools in the middle of the last millennium BC runs parallel with agricultural and demographic expansion in the European interior. Evidence is provided not only by the great size of late Iron Age hillforts and other settlements, but also by the huge armies that began raiding and invading the Mediterranean world from the fourth century BC. Greeks and Roman were terrified by these invasions, but had no real notion of their causes.

The temperate Europe that Julius Caesar and his successors found when they began serious warfare north of the Alps was already tamed. There were no longer any hunter-gatherer populations. Not was there any primeval forest. Forests had expanded in the early Holocene, following the retreat of the glaciers, but they had mostly been felled by the first farmers. The woodlands that replaced them were created and managed by human activity. Roman poets wrote of northern Europe as utterly savage, a continent of dense forests full of wild beasts. But Roman generals, and the tax collectors that followed in the early empire, will have appreciated the phenomenal agricultural productivity of European landscapes compared to the arid Mediterranean. Forests and game certainly survived, as they do today, but only on the higher ground between cultivated landscapes. Around the Mediterranean there was already the present-day landscape of garrigues, a characteristic type of shrubland rather like chapparal formed of plants that can tolerate the summer heat, alternating with small cleared plains on which cereals might be grown.

Prehistoric agriculture was not limited to the production of grains: woodlands and wetlands were also exploited; salt, vital for preserving meat and fish, was mined, gathered from coastal salt pans, and traded; large herds of livestock were raised, and short-range transhumance was already practised. The range of animal and plant species cultivated by prehistoric societies might seem relatively small, but the impact of farming on the early Holocene fauna was already phenomenal. The top predators who had expanded out of ice age refugia into Europe now diminished in number as their prey and habitats were removed. Lions disappeared from Europe and eventually western Asia, while populations of wolves and bears were fragmented. Greek heroes fought with savage beasts that threatened grain-growing lowlands, monsters like the Nemean Lion and the Calydonian Boar. Those myths reveal a world already imagined in terms of a stark opposition between civilization and the wilderness. And the wilderness was in retreat. Over the Iron Age and Roman centuries, smaller domesticated cattle replaced aurochs in temperate Europe, and bison and elk were restricted to the far north. Deer retreated into the remnant forests and resurgent woodlands. Hunting became rarer and less exciting. When the Emperor Trajan wanted to hunt game in Italy, he had to climb to the summits of the Abruzzi. The Roman generals who conquered the east were amazed to find Hellenistic kings had, in imitation of the Persian emperors, created reserves in order to preserve animals worthy of a royal hunt. The Persian word was paradeisos, giving us our term ‘paradise’. Wilderness had become a scarce commodity, a luxury that needed to be conserved and cultivated. We can easily imagine this today, just as we can easily imagine the rich rewards that would follow Roman possession of the waking giant of temperate Europe.

Ecology and Empire

Imperial expansion since the fifteenth century has often had dire environmental consequences. One reason is that globalizing movements often connected up regions that had been out of contact for long periods. Most dramatic was the Columbian Exchange of plant and animal species that followed European discovery of the Americas, leading to extinctions and the catastrophic merging of disease pools, as well as the transformation of Old World diets with the introduction of coffee and chocolate, potatoes and cane sugar.¹²Then there have been deliberate modifications of colonial ecosystems in modern times, such as the creation of cotton and coffee plantations in the Americas, the introduction of cattle ranching into parts of North and South America, of sheep farming into Australasia, and the transplantation of maize from the New World into Africa. Cash crops often replaced subsistence agriculture, and the needs of distant imperial markets took priority over those of indigenous populations who were sometimes dispossessed, and sometimes conscripted to labour within new agricultural regimes. Slavery notoriously allowed the wholesale movement of human populations. There are some prehistoric precedents. Human expansion at the end of the Pleistocene era, into first Australia and later the Americas, seems to have led to the extinction of native megafauna, from two-tonne giant wombats and three-metre-tall kangaroos to the American lion and giant sloths. The settlement of the Pacific Islands and New Zealand was accomplished only because explorers took pigs, chickens, dogs, and various domesticated crops with them in their canoes.

Roman expansion did not have such dramatic effects. The empire expanded within a region the inhabitants of which used broadly the same domesticated species as themselves. When areas beyond the Mediterranean were eventually incorporated into the empire, this was usually just the latest phase in long histories of contact. As a result, Romans rarely encountered economies or ecologies very different from their own. The environmental changes that Roman expansion brought were, on the whole, more piecemeal and more subtle than those introduced by European empires.

All this marks an important difference between the ecologies of modern and ancient empires. Roman expansion was facilitated by what the conquerors shared with their new subjects. The first tax levied (on Sicily) was a simple tithe of grain, and when armies campaigned in Spain in the second century or Gaul in the first century BC, their local allies were expected to provide food. African Lepcis paid a great indemnity in olive oil, and the Frisians at the Rhine mouth were taxed in hides. Taxation in kind was always an important part of Roman fiscal systems. Yet even when cash was required provincials could earn it simply by intensifying the production of crops they already grew, since these were the crops Romans already knew and desired. Monocultures and cash crops never squeezed out pre-existing regimes in the Roman period. We know of no disastrous experiments with new crops. The empire expanded into regions that were already productive, and in most areas did no more than stimulate a modest intensification.

The political unification of the Mediterranean has even been seen as simply the latest—political—phase in a much longer story, one that had begun with the spread of farming out from the Near East. It has been suggested that the limits of Roman expansion were set, in some regions at least, by the limits of this Old World agricultural complex. To be sure there was no ecological frontier dividing Roman from Persian spheres of influence, indeed that frontier cut through areas in north Syria that in every possible respect—cultural, religious, technological, and ecological—formed a unity. Yet Rome’s northern frontier at points reached areas where the returns from this style of agriculture do not seem to have been impressive during the Iron Age. Perhaps Roman conquest reached an ecological limit in northern Britain or the Low Countries.¹³ Most of the long European frontier ran through rich agricultural land. Where Roman rule did meet real ecological limits was the Atlantic and the Sahara. The empire had successfully gained control of the southern part of that western corridor leading out of the Near East: threats came to it from north and east.

Roman expansion followed agricultural change, and brought with it few new technologies, species, or diseases: consequently, Roman conquest had no cataclysmic ecological impact. But this does not mean that Romans were not positively interested in promoting intensification. One sign of Roman interest is the care they took to learn from the agricultural regimes they incorporated. The only book the Romans took from the libraries of Carthage after the sack of the city in 146 BC was Mago’s agricultural treatise which was then translated into Latin. Generals returning from the east brought back new species of plants, including cherry trees. Roman writers of the early empire were actually very well aware of how recently many nuts and fruit crops had been brought to Italy from the east. More than 40 per cent of the plants named in Columella’s first-century AD book On Agriculture were Greek in origin. The Natural History of his near contemporary Pliny the Elder describes in detail the various trees and other crops found in different parts of the empire, and their nutritional and sometimes medical benefits. Medical texts produce a wealth of information about cultivable species. Pliny was just as interested in the plants and animals of the western Mediterranean and even temperate Europe, but it is obvious from his account that the main direction of movement remained east to west.

Roman entrepreneurs not only brought eastern crops to Italy, but also attempted to transplant various Mediterranean species north of the Alps. Apples, pears, cherries, plums, and walnuts were introduced to northern Europe very early on, as were celery, garlic, asparagus, cabbage, and carrots. Chestnut, probably used for wood rather than primarily as a food source, followed a little later. What these trees and vegetables had in common—and what separated them from cereals—is that they required cultivation in gardens or orchards. Arboriculture involved a range of new specialized skills, such as grafting. It also required much greater inputs of energy and time in return for greater calorific and financial returns per hectare. The northward progress of these crops—like that of the vine—reflects the spread of Mediterranean taste, as well as of agricultural knowledge. Carbonized remains of these crops appear on settlement sites alongside those of others that could not be domesticated north of the Alps including figs, chickpeas, pistachios, almonds, pine kernels, and melons. The expansion of arboriculture and gardening was also closely linked to urbanization. Not only is it labour intensive but it also relies on levels of consumer demand most easily found where populations are densely packed and relatively prosperous. It also offers vital supplements to a diet high in carbohydrates, providing sugar, protein, and vitamins. It was urban populations that had the greatest needs of these supplements.¹⁴ It is no accident we know most about it from Roman Egypt and the environs of Rome, the most urbanized parts of the ancient world.

More generally, the spread of arboriculture allows us to observe Romans as enthusiastic adopters of new cultigens. Profit and the desire for a wider range of tastes were no doubt key motivators. But there was a sound ecological logic too. The principal weakness of the earliest agricultural regimes in all parts of the world was the small number of cultigens on which each depended. Dependence on a single crop is enormously risky, and diversification provided a key buffer against crop failure. Hence the early importance of pulses, and the steady growth of the variety of grains cultivated, naked wheats joining emmer and einkorn alongside spelt and barley, and rye, oats, and millet being grown in some regions. The wider range of cereals did not only provide better matches for particular local microclimates. Barley was less favoured as a food but could resist drought, millet could be grown over the summer, chestnuts and other nuts provided a key protein source when crops failed, or fodder for pigs when they did not. The range of cultigens increased over time.

Other ecological niches were subjected to the same process of intensification: wetlands and mountains, woodlands and the pre-desert. Roman construction technology allowed a few innovations here, most notably in the imperial period. Injections of capital are difficult to document, but they must lie behind some agricultural development such as the construction of massive sheep pens on the plain of the Crau in south-east France, and the development of industrial-scale production of fish-sauces. A small amount of irrigation farming was introduced, a spin-off of the aqueducts bringing water from highland areas down to the cities of the plain.¹⁵ Terraces were constructed across seasonal river courses in the Libyan pre-desert to catch floodwater. Drainage of marshes took place across the empire from central Italy to the English Fenlands. Hydraulic engineering allowed for advances in fish-farming. Establishments dedicated to salting and pickling fish appeared wherever catches were abundant, including on both sides of the straits of Gibraltar, up the Atlantic coasts of Spain and Gaul. The processing, storage, and transportation of agricultural produce was in general improved. Large mills driven by water, animals, and slave labour supplemented hand mills, and large-scale olive presses were constructed.¹⁶ Advances in granary construction and harbour installations went hand in hand with the manufacture of earthenware containers and ships better adapted to the transport of bulky produce. Road infrastructure was improved, perhaps first for military reasons, but others benefited too. Closely allied to this was increased production in the most common urban productions, such as textile production.¹⁷

A different form of intensification is evident in stock-raising. The scale of sheep farming increased in several parts of the empire, including Italy and south-east Gaul where very large flocks clearly catered to demand from outside the region. There were highly successful efforts made to increase the size of the main meat-producing domesticates: the growth of cattle of various breeds is now well documented from faunal remains in Italy. North of the Alps it occurred so rapidly that it seems almost certain that new breeds were introduced. This reversed a long trend in Europe towards smaller and smaller animals, showing very clearly the impact of new priorities and techniques.

The cumulative impact of these improvements was economically significant. Appreciating them helps us understand how the Roman Empire sustained, in such an unpromising environment as the Mediterranean basin, a ratio of consumers to producers that rose by the early third century from one in ten to three in ten in some regions. City life brought a demand for bread instead of porridge and for a more varied diet. The archaeologically visible result was a proliferation of bakeries and food markets, macella, that sold fresh meat, vegetables, and dairy products. Improved communications and the muscle of civic and imperial authorities and of landlords made such a diet possible for a minority. But the long-term environmental impact was limited. When the cities (and so their aggregate demand) shrank, and when authorities could no longer maintain roads, aqueducts, and the like, rural economies shifted back to a more local scale. The Roman urban boom left few lasting environmental traces, except in respect of mining, which generated levels of heavy-metal pollution that would not be reached again until the Industrial Revolution. Repeated attempts to convict Roman civilization of causing deforestation and soil erosion have failed to convince. Roman expansion led to an intensification of production, not the wholesale transformation of their environment. Their imperial ecology was very different from that of the modern age.

Further Reading

The most influential environmentally organized account of antiquity is Peregrine Horden and Nicholas Purcell’s The Corrupting Sea: A Study of Mediterranean History (Oxford, 2000), which has already prompted many responses, some of which are gathered in William Harris’s Rethinking the Mediterranean (Oxford, 2005). Equally innovative is Robert Sallares’s The Ecology of the Ancient Greek World (London, 1991). Conditions on the Steppe are discussed in Roger Batty’s Rome and the Nomads (Oxford, 2007). Brent Shaw’s essays on North Africa are collected in Environment and Society in Roman North Africa (Aldershot, 1995). No environmentally oriented account of Roman temperate Europe has yet been produced, but there is much of relevance in Chris Wickham’sFraming the Early Middle Ages (Oxford, 2005).

A realization of quite how precarious conditions could be in parts of the ancient Mediterranean is relatively recent. A pioneering collection of answers to the question of how ancient farmers managed was Paul Halstead and John O’Shea’s Bad Year Economics(Cambridge, 1989). On the means through which classical civilization was sustained in the face of these stresses the fundamental work is Peter Garnsey’s Famine and Food Supply in the Greco-Roman World (Cambridge, 1988).

The impact of Roman agriculture and especially mining on the environment is a topic of current debate. A good starting point is the collection edited by Graham Shipley and John Salmon, Human Landscapes in Classical Antiquity (London, 1996). A vivid account of the environmental cost of ancient mining is presented in chapter 7 of David Mattingly’s Imperialism, Power and Identity (Princeton, 2011). Two very well-written introductions to the prehistoric background are Tim Champion et al., Prehistoric Europe(London, 1984) and Graeme Barker’s Prehistoric Farming in Europe (Cambridge, 1985).