Friday, May 13, 2016

A Wizard Did It! The Fall of Rome, VI

Is the fall of the Roman Empire all down to monetary policy? Given Aurelian's precedent-setting method of dealing with central bank independence, it's kind of relevant right now. 

Not Aurelian and Felicissimus.

It's also top of mind, in that Brad DeLong is on about Marius and his war of extermination against Cimbri and Teutones. Marius and the Cimbri are, indeed, a huge moment in history. Marius may or may not have destabilised the Republic and changed its army. (Emphasis on "not.") He most certainly served as a patron to his brother-and-law, and promoted his nephew's career insofar as he could, and since that nephew was Julius Caesar, Marius is one of the progenitors of the Roman Empire.

I am, however, taken with the intersection of geography with our long struggle with the past and its constant reinterpretations, and drawn to alternative vectors of work and life. Marius fought Gauls; Romans and Gauls have always been fighting. Through centuries of reflection and recursion, it is not at all clear whether the first war of 390 BC, or the last war of Augustus (the context, after all, of the author of Livy's History, which tells us most of what we know about the previous ones) should have precedence. What we do know is geography and lifeways. Somehow, those wars took Roman walls and Roman forts, Roman meadows and Roman fields, all the way from Latium to the Rhine, and that last step proved one too far, so that we may well ask what  happened between Marius and Augustus, what went wrong, if that is when it went wrong? (A hard hill to climb, given a half-millennium of Roman success afterwards. But still. . . )

It's just, and I don't want to seem churlish towards the guy who hosts the great website, but I just don't see it. I mean, I see wars against Cimbri and Teutones; I just don't see them as reactive wars of defence. Rome and Gallic wars are a problem for historians, and should be a problem. The first Roman-Gallic war is quite clearly the last. Livy, court historian of Augustus, who sent his legions north to reconquer Gaul and establish themselves on the Rhine, is confessedly the inventor of that first war of 390BC. He thinks he has good reason to suppose that it happened, but he is naive about these things. The wars of the third century are similarly vague. I mean, there were actual wars, against actual tribes and actual places, but their transformation into a "Gallic" barbarian threat against Rome smacks of "me-tooism" coming out of the recent attack on Delphi.

 Frankly, with Marius, we get the the idea of a Roman state faced with an existential threat that demands a defensive campaign against an enemy host on the far side of not merely the Appenines, but the Alps! This. Is. Cuckoo. No. Really. This is bold all-caps crazy talk. Hannibal notwithstanding, armies quite simply do not, before the Napoleonic era, cross the Alps in fighting order. They arrange for allies on the Italian side to receive their columns and provide them with quarters where they can restore themselves. A coherent bodies of warriors and their dependents who have migrated from Denmark to the Alps, and there linger for a decade awaiting their opportunity to descend on Rome --that is not how barbarians work! And, if not barbarians, then what the fuck are we talking about? Ten years is a long time in the life of a young cowboy. Fire in his belly in 113, perhaps, but wife, child and farm by 101, and so much for adventures across the Alps. If "Cimbri" and "Teutone" there were in 101, as in 113, then it is because of a sociopolitical continuity that defies everything we are told about them to begin with!

In fact, It seems much more plausible that the story of Marius is the story of his nephew. Does anyone take literally the idea that Julius Caesar and August were waging "just," defensive wars in Gaul? Of course not! So forget the old bullshit. It suffices that an ambitious politician, shut out of other, lucrative colonising games, turns an avaricious eye northwards towards the line of friable soils that lie along the spring lines of the hills that flank the great subcline. Everything else is moonshine. Well, not everything else. He needed campaigns, and loot, to satisfy his veterans, along the way to centuriating Bologna, and many other colonies, too. This new region of Roman settlement would from then on be an important base of support for him, and, in time, his nephew. To fill in the blanks, I suggest that wherever Marius' far-flung campaigns (looting expeditions) took his troops, the region that he consolidated and shared out was his rear area. Ask postwar Japan and Korea about what comes of being the zone of communications of an imperial army at war!  

From the first, then, money.

Monetary issues have already been suggested, as the "traditional" explanation of the fall of the Empire,* that the collapse was to some extent monetary. Various villains may be fingered, but it is down to Roman emperors after Septimius Severus raising the army's pay, and someone meeting the burden by issuing silver denarius coins with reduced amounts of bullion rather than by obtaining more silver. This led to runaway inflation and a near complete collapse, until Diocletian and Constantine came along and fixed it in ways that made it worse in the long run. The suggestion is that the only way to get a "centrally planned" economy to work is by introducing a Dominate, which is totalitarianism with a Roman face. Sounds scary, and, of course, was what the kids are calling a "hot take" back in 1952. 

When (second-to) last I left this topic, it was with this observation, followed by the suggestion that the details more obviously suggest deflation than inflation: too few coins issued, and not too many. (In other words, I'm no better than those old John Birchers. I'm just using a current hot take.) Still, it's a marvelously economical explanation, if you don't believe in barbarian invasions. The walls that sprung up around Roman cities in the West, first in Britain in the second century, then in France and even around Rome, can be explained as attempts to enforce customs barriers and extract tax revenues from market goods. The unrest against which Aurelian was guarding becomes endogenous tax revolts rather than exogenous invasions with the considerable obstacle of the Alps and Apennines to overcome before arriving at the gates of Rome,

The corollary, then, is that while successful at first, the longer term result was the decline of the Roman civitates in the West. This is something that we have to account for anyway, and the tax walls would have killed the official markets through competition with informal cross-roads markets where people exchanged tally sticks. (Etc, etc.) Roman monetary policy is too tight, and it kills the state. (Conclusion first: now I only need to work the evidence to serve!)

Image scraped from Bonnie Bishop-Dudasik's British Museum-associated Pinterest site. 

This has an interesting corollary, drawn out for me recently by Patrick Geary: a lack of big capitalists. While not short of wealth, or of loansharks, the late Antique Roman senatorial class did not issue loans to the emperors, who instead just issued job lots of low-silver coins when they needed to mobilise against an usurper. Finance, like technology, was not up to the job. Why  it wasn't is probably not on the table here, since this seems like something that could have been done under the Republic. 

This is, in a sense, an old argument about the Roman economy, which was not as innovative as some would like. Sidestepping the old history of technology arguments, which leave me distinctly dissastisfied, I have preferred to point to a curious lack of extensive development on the peripheries. On Iceland and in the Faeroes, in the high Alps, the Sahara, Macronesia, and in Scandinavia and Ireland, a long list of regions immediately outside the Empire which could (ought?) to have developed rapidly through trade with Rome, quite simply failed to do so. The fact that many of these regions began trajectories of economic development in the immediate sub-Roman period suggests that the Empire was a drag, rather than the reverse. 

We have, after all, the strange case of an economy which is short of bullion, yet cannot be bothered with trans-Saharan gold, which Africans will all but force on you if you can just get there to trade. It's not like Africans had to wait to invent camels! (Though the first source I turned up on Google wants to make it all dependent on this technological exogeneity.) 

It is odd, and there is a case for making it global. The Han fell synchronously with the rise of the Severans and the Wen Emperor restored unified rule in 580, just as, in the west, the "Avars and Slavs" were supposedly rewriting the human geography of Greece. (I don't doubt the rewriting, just question the agency.) While this new geographic order persisted in Greece into the Nineteenth Century and so can be studied, there is at least room to wonder whether the same thing happened in Italy at the same time, with the rise of the Lombard Duchies of Spoleto and Benevento

By Italie_1000_AD.svg: * Original author : MapMasterFrench translation by : CyberproutModified by : Ewan ar Born and Sémhurderivative work: Demuro (talk) - Italie_1000_AD.svg, CC BY-SA 3.0,

(So also note this:)

By Italy_400bC_It.svg: Decanderivative work: Richardprins (talk) - Italy_400bC_It.svg, CC BY-SA 3.0,

I've just heroically conflated the "sub-Roman" with the height of the Republic. Historical inquiry. How does it work? All I can say is that we're laying off a vast and profound historical event to monetary policy, and monetary policy is as close as we currently have to magic. A wizard did it! 

So, like I said, it's a theory. Is there any progress from the point of view of the kind of evidence that I want to bring to the table? I'm glad you asked!

Zurich's Lindenhof Hill from across the Limmat River, painted in the late fifteenth century.

A second-century tombstone found at Zurich identifies the Swiss city at the internal delta where the Limmat River flows out of the Zurichsee towards  the Rhine as "Turicum Station of the Fortieth of the Gauls." What, precisely that means, we do not know, as we have no literary source for the Quadrigesima Galliarum. We do know enough that it was a line of customs stations, operated by tax farmers, built on the major routes into and out of Gaul. It was intended to levy a one-fortieth customs due, and might have been conceived as an Italian tax on Gallic commerce. Although, that being said, four have been identified in Switzerland, for example, and another in the Pyrenees. I am not here to teach the Romans geography, but Zurich at least is not really positioned to tax Gaul's trade with Italy. It might arguably be for taxing Cisalpine Gaul's commerce with Germany, or Gallic trade with Raetia, but rather than stretching one's speculative muscles in those directions, it is probablly better to settle for the idea that it reflects the geographical understandings of either its likely founder, Augustus, or a later successor such as Hadrian, as Turicum station itself appears to have been built after 75AD, and the adjacent temples to Diana and Sylvanus were built in Hadrian's time.

The tax farmers of the qG were replaced by procurators under Septimus Severus, and a Valentinian fortress was built over the Turicum station sometime in the 370s. What, if anything, that means for the late Roman economy is a mystery in the face of our scanty sources. I am just making a big deal of it because Roman activity in the high Alps is otherwise so evanescent. Given the extent of trans-Alpine trade between southern Germany and northern Italy, this is a Fact of Interest. It's also hardly the only example of a peculiar slackening of economic activity at the margins of the Roman Empire --and well beyond.

From here I could go to a lot of places --England, Denmark, Iceland and Hawaii all have their charms, but this time, I think, Nevada!

Welcome to the Great American Desert, in this case the Timber Creek country in the Schell Mountains in northeastern Nevada. 

The comparison I am going for here isthe xeric woodlands zone of the Sahara.  For a fairly relaxed definition of forest, these uplands are forested, too. The higher, the more humid, but also colder and the shorter the growing season, which is why, when the European explorers finally did reach the mountains of the Air and the Agadar, they did not find ancient cities of red men with flying ships and atmosphere plants.

Closer to sea level, Nevada produces more stereotypical landscapes.
Valley of Fire State Park. 
Now we just need a roadrunner. Now for another take on the Nevada desert: Ruby Lake National Wildlife Refuge. I feel bad about scraping this image from a commercial site, but it gives a good visual sense of this ecosystem:

Ruby Lake is a wild life refuge because it supports antelope and mule deer. It does not support elephants, but mammoths have been found in the Nevada desert. This might be down to climate change, but it is likely that, on the basis of existing elephant populations, that areas like Ruby Lake could have supported endemic mammoth populations even if the weather here as no wetter when North American mammoth populations became extinct.even if the weather here as no wetter when North American mammoth populations became extinct. 

Why does this matter? Because we know that the Romans sought charismatic African megafauna for arena games --and, before that, there was this thing with Hannibal and elephants, maybe you've heard of it? 
J. M. W. Turner, Hannival Crossing the Alps. Source.  Although the Tate Gallery will let me license this 213-year-old image for a very reasonable fee.

African elephants (forest elephants, that is, the so-called "Indian" elephant) are, we know, found south of the Sahara. This has led to considerable archaeological effort to find evidence of Roman activity south of the Sahara, effort which has proven futile.  The Romans interacted with a state based in the Fezzan on a regular basis, and a major traditional trans-Saharan trade route leads from Fezzan to the Tibesti Mountains and on to Lake Chad. There are even grounds to argue from Roman geographers that the Garamantes of Fezzan were active in the Tibesti. It's just that there are no Roman trade goods south of the desert.  At all. Then comes the end of the Romans, a hiatus of uncertain period in which strange things happen, and then, definitely, trans-Saharan trade!

The American evidence suggests an alternative: before the spread of  irrigation,presumably from the Fezzan, what are now agricultural oasses were desert wetlands occupied by a mix of pastoralists and megafauna. The Romans (and Carthaginians) did not have to cross the Sahara to get elephants. That does raise the question of why foggaras were not built in other likely sites before sub-Roman times, but it seems like a chicken-and-egg question given the importance of the trans-Saharan trade to the oasis economy. If Rome is failing in its role as driver of long distance trans-Saharan trade, there is no particular reason for indigenous populations to invest in capital-intensive irrigation tunnels.

We're left, then, with a Roman frontier with a customs boundary, and a strange lack of economy activity. Something doesn't add up. Now I want to make an effort to see if it does add up. How far does the African analogy help us with the Alps?

I'll start with another historian obsessed with the way that recursion makes it impossible to say which comes first. Alexandre Grandazzi(1) begins with the mulitple origins of Rome: as Greek, Etruscan, Latin and Trojan city, in approximately that order, and excavates strata of ancient narratives, beginning with the lacunae of 390BC, where Livy finds his first written sources, and thus locates his Gallic invasion, perhaps for no better reason than to justify his patron's wars, perhaps on the basis of early narratives of the third century; but certainly not because there was then, or now, evidence that Rome was actually sacked by any Gallic invasion in 390. 

Sorting out the narratives into their proper sequence and hermeneutically extracting history from them would be an impossible task without archaeology as a guide, and much of Grandazzi is a discussion of the excavations at Rome, justifying the idea that a city really was quite self-consciously erected there in about 730BC. The glamour of far Nineveh, and the glory that was Assyria will serve as ideological justification, I propose, while Grandazzi provides a much more useful economic substructure. 

First, we know what that substructure was not. Latium is a land of hills with interstitial lowlands, walled off from the sea by the Pontine Marshes, which is in turn impounded by a line of sand dunes on a Pliocene horst sill. The canals which drain the marshes and wetland which form on the impervious bedrock under  the volcanic basins of Latium, were cut in the fifth and fourth centuries BC. Before that, the peasants cultivated the friable land on the slopes of the hills, and the spring line was the axis of settlement, and the interior wetlands, like the vast expanse of the Pontine Marshes and the salt marshes at the mouth of the Tiber were for grazing. All of these animals demanded salt, and the salt industry was the primordial cause of Rome's rise. Not a saltern itself, it was the place to which all the roads ran, as they say. This was not because of empire, although Grandazzi makes heavy weather of justifying the historical record of the regional hegemony of archaic, royal Rome; but rather because the roads must come down the Tiber, or skirt the Pontine Marshes until they break through to the sea along the Tiber. Sabine, Sarmatian, Umbrian and Oscan, all must come down to the sea via Rome, and buy salt with the increase of their herds. It is the rise of the Latin cities, later, which challenges Rome and forces it to fight for its lost status of regional hegemon in the fourth century, a fight in which it is so far victorious by the early third century that it can begin to look further afield. 
Ruins of Alba Longa: Source. Note that if the ruins of Alba Longa actually existed, archaeologists and historians would have to find something else to fight over.

If history repeats itself, we know the next stage: a confrontation between Rome and the upland herders along the line of the Via Flaminia stretching to the Adriatic at the mouth of the vast subcline of northern Italy. This is the multi-sided war vaguely sketched for us by the Liber Pontificalis in the early Medieval period, and, perhaps, the structure on which we can hang the vaguely-known wars of the third century BC. It is, therefore, with some pleasure that I can point to a worthy master's dissertation turned into a fine book. Jeremiah B. McCall[2] gives us a history of the Roman cavalry that goes beyond Nineteenth Century controversies between cavalry and "mounted riflemen." As he tells us, in the earliest days of the Republic, the elite Romans who voted in the prestigious eighteen equestrian centuries received horses subsidised by the state and served in the army as equites equo publico. (It is hardly an accident that eight equites are drawn from the nobility of the former Alba Longa, incorporated into Rome as the inhabitants of the Caelian Hill by Hostilius Tullus.) 

This narrative focusses on military service as an extension of social privilege. You can be Roman cavalry because you are sufficiently well born, and as if this bit of conscious elitism were not enough, wealthy citizens who could afford to provide their own horses could fight as equites equis suis. McCall's decisive intervention is to show that equites equis suis must have been mobilised fairly often, contrary to some opinion, and the key lever  that opens the floodgates is, of course, the Tumulto Gallica, the state of total mobilisation against the Gallic threat which is enthroned as a constitutional provision (so that it can be invoked against Pyrrhus and Hannibal) and used as a pretext to put every man who can be put astride a horse in the field.

And from where? After Grandazzi's grand and characterisically Gallo-Italic claim to be the pioneer of something called "historiology," and the slightly suspicious prestige of a junior historian writing on subjects military, it is a relief to turn to a massively empirical Anglo-Saxon regional study of Umbria, by Guy Bradley.[3]

Wow. By The original uploader was Wetman at English Wikipedia - Transferred from en.wikipedia to Commons., CC BY 2.0,

After the obligatory survey of the historiography of the Umbrians (pretty scant considering the size of the region and its historical importance) we begin our approach to their history. Earliest sources (Herodotus, Strabo and whoever he is quoting) have them spreading from historic Umbria to the headwaters of the Danube, contesting Rimini and Ravenna with incoming Etruscans. The Greeks, Bradley speculates, knew the Umbrians as the coastal people of the Adriatic coast of Italy, which is why we have scattered mentions in Herodotus and Aristotle (the Meterologica has a fascinating fact about the Umbrians making salt by casting ash into water. Salt, charcoal and iron!) Stabilisation and stadialisation occur in the Final Bronze Age and early Iron Age, with burials showing the occupation of the pre-Roman cities in the fifth and fourth centuries. Again, the archaic sites are on the springline below the 500 meter contour, at the boundary (dare I say pomerium?) between agricultural plain and wooded, pastured upland, although the lowland marshes were drained by the Romans, from about 230BC, much later than the ones of Latium. Exploitation was not predominantly pastoral, but some kind of animal-based exploitation of the  uplands could hardly be avoided in a region that contains the main spine of the Appenines, with much of the land above 500 meters in spite of containing the easiest passes of the region. The spread of villas at the expense of small farmers marks the second century BC in Umbria as well as Roman territories.

On the basis of the only full field survey in Umbria, taken in the basin of Gubbio. This shows a substantial influx or intensification in the middle of the first century BC, marked by the use of Roman Republic black-slip pottery. However, vegetation analysis suggests a much more gradual intensification of agricultural use of the region going back to the early Iron Age. It is the imported pottery that is at issue, and it is best interpreted as evidence that the region around Iguvium was becoming more implicated in long-range trade in this period than before –or, apparently, after. Along with intensification in the valley bottoms, there must have been intensification in the pastures above, for they were rich. Cicero, De Div, notes that the Umbrians are as skilled at augury as the Arabians, Cilicians and Phrygians, as they, like them, were “chiefly engaged in the rearing of cattle, and so they are constantly wandering over the plains and mountains in winter and summer.”

In short, by the late Republic, northern Umbria is showing signs of specialising as  a cattle-raising country. This would not surprise Luk de Ligt, whose intervention in the contentious field of ancient Roman demography insists on land-use change.[4]. The natural extrapolation from the "common sense" geographical distribution of market-oriented agricultural production is that a peripheral zone of intensive stock raising, attested in Italy for Apulia and Lucania would provide the store cattle for feedlots in the immediate vicinity of Rome, which would support the kind of agriculture attested there: orchards, and vegetable gardening (and dairy and poultry) would all have required plenty of manure and produced plenty of feed. The missing input, not attested from Umbria but kind of famous in other Roman contexts, would have been grain from "wheat bonanza" lands, as they said in Victorian times.

There remains, then, only the question of funding the state: and that's what the Roman walls are for! And so we have a historic economic model. Can it be extended from the valley of the Po to the valley of the Rhine, as Augustus supposed? Perhaps this is a bridge too far, but if sheep raised in eastern Oregon can be sold to meat packers in Chicago even before the railroad, perhaps not. 

And then, in 75AD, the customs line of Gaul sprouts a new perimeter, along the mouths of the passes of the central Alps. As Tacitus says, the secret of empire is that an emperor can be made outside Rome. Whatever he means by that, you can now farm taxes in the high Alps. That is, until the Severans get rid of the tax farmers, which seems like an odd choice. Since, after all, who better to borrow from when an army must be raised, then from the men through whom all the money must pass?

 Unless there is something wrong with the money. We can --vaguely-- guess the logic. Buy silver at a high enough price, and silver will be mined. buy it with what, though? Tax remittances or with gold, I suppose. Africa will serve as evidence that there is no sufficient desire on the part of the Roman state to stimulate trade to get gold where gold can be had. Investment is failing, not reaching where all economic logic says it must go. Rome never reached trans-Saharan Africa. Now it is going to give up Germany, if the only alternative is to just start issuing a silver coinage in sufficient quantity. 


*Hunter Marriner Mountain Strawman, The American Historical Assiociation Presidential Address, 1929: The Roman Empire Fell Because of Those Damn Demmiecrats Now Let's Adjourn to the Billiards Room For Postprandial Cigars)

1. The Foundation of Rome: Myth and History Trans. Jane Marie Todd (Ithaca: Cornell University Press, 1999)

2.  The Cavalry of the Roman Republic: Cavalry Combat and Elite Reputations in the middle and late Republic (London and New York: Routledge,. 2002) 

3. Ancient Umbria: State, Culture, and Identity in central Italy from the Iron Age to the Augustan Era (Oxford: OUP, 2000)

4. Peasants, Citizens and Soldiers: Studies in the Demographic History of Roman Italy, 225BC—AD 100 (Cambridge: CUP, 2012)


  1. I'm up for the idea that "before the spread of irrigation, presumably from the Fezzan, what are now agricultural oases were desert wetlands occupied by a mix of pastoralists and megafauna". The example of Namibia today would seem to suggest that elephants might have been capable of surviving around the oases, too. But if we may believe Pliny, elephants were still found throughout modern-day Morocco in his time - he mentions them around Sala (Salé), Mount Atlas (the Moroccan Atlas, that is), in Tingitania (Tangier region) and the Seven Brothers (Ceuta). He even mentions them in Syrtis, though not in the more settled areas of Tunisia and Algeria. So initially the Romans don't even need to get as far as the oases to obtain their elephants - or at least not until they start running out of Moroccan elephants. I don't know when that would have happened exactly, but let's just say I don't recall any Arabic geographies mentioning elephants in North Africa... So you could maybe make the extinction of Moroccan elephants drive the early stages of economic expansion into the Sahara. Perhaps we can envisage nomads capturing elephants and selling them, then investing the proceeds in foggaras and slaves to run them, thus destroying the ecosystem configuration that made elephants available, and finding themselves impelled to explore longer-distance trade - new polities coalesce around the elephant hunt, and end up taking over the steppes to their north*... Still, a few suitably dated Saharan elephant bones would be nice.

    * Fentress and Wilson, now out:

  2. Thanks for that, Lameen. I am thinking I need something lapidary: Changing economic geographies as evidence of contractionary Roman monetary policy. It's just that while I have an excellent feed on the Sahara, I need to actually get to the library to talk about the Alps.

    Or just noodle around on Google:

    I'm also thinking about Spoleto as a triplet of Tripolis and Constantine, but that's probably too much for one post.