Fall of Rome, A Recap, One: Honorius tells England That He's Invoking Article 50: Some Underappreciated but Scarcely Unknown Economic Aspects of the Problem. You'd Think.

So you might have expected the second part of June techblogging about now. Up until 12:16, Tuesday afternoon, so did I. That's when I was asked to give up my Victoria Day statutory holiday because my company can't find someone else to do my job in the city of Vancouver for one night. For those funny-looking foreigners who don't celebrate the name day of Victoria of House Saxe-Cobourg-Gotha, it's on May 24th. Don't worry, though. I get my May 24th statutory holiday next week! You know, after we celebrate Canada Day. As for my Canada Day statutory holiday --further bulletins as events warrant.

As I have said before, you would think that this would reflect a "labour shortage," in that there is a shortage of "labour" to do the job. (Lots of jobs; not just mine.) You would, however, be wrong. In so far as I understand this "economics" thing, shortages cause prices to go up, and the price of (my) labour in this city is not going up. It's going down. 

I know, blah blah globalisation. But you can't outsource a neighbourhood grocery store, and we already hire every immigrant we can. Something else is going on, and I am not sure how well we understand it. 

The bright side of this is that I get to link a post about the distant past to current events. It's a hot take! The disadvantage is that it's hard to write about what you don't know, if you don't know what you don't know. (Or you'd think it ought to be hard.)

I, for example, do not know why the house at 7249 Cartier Street, theoretically worth seven figures, is sitting abandoned, and looks like it was abandoned more than ten years ago, well before the current "red hot housing market" started --perhas going back to the last boom. Maybe they have fled the oppressions of modern life to live up some BC sideroad! Haven't we all dreamed of "going off the grid" that way?

Much more likely, the owner went "off the grid," by going into a nursing home. There's lots of ways of going off the grid, including not existing; but, also, from the point of view of the economy, by not working, or not working very hard.

Honorius, proclaimed Imperator, semper Augustus, to the cities of Britain: Fuck Off And Die. "Emperor Honorius," by Jean-Paul Laurens (1880). Image from Wikipedia, inspiration from Laurens' "opposition to monarchical and clerical oppression." The only people Honorius looks like he oppresses in this picture are PvP opponents with female nyms in Final Fantasy chatrooms, but whatevs.

The other thing going on here is a top-voted response on Quora explaining why the Roman innovation of a standing army was not repeated in later historical epochs. (It's not top-voted any more, but it was when I saw it.) Stephen Tempest explains that ancient regimes had "90%" of their population involved in agriculture, with the residual 10% involved in everything else besides, and that this 10% of the population had to supply legionaries as well as millers, butchers, bakers, and telephone sanitisers. This is wrong, but it's wrong in an interesting way that justifies a bit of a data dump here.

So, anyway, Emperor Honorius didn't actually tell the civitates of Britain that he was putting Roman Britain into liquidation, and that the garage sale would be Sunday. He just told them to look to their own defence. This was something that Theodosian emperors said to many cities, many times, and there's even some question as to whether the famous "Letter to the Britons" might actually be a letter to Brutium. Still, this is the whole "end of the standing army" for Britain, at least.

It's also reasonably likely that he intended this to be read aloud in a very sarcastic tone. The Roman garrison in Britain had been proclaiming and supporting usurpers practically nonstop for years. Yes, the cities of Britain had been kind enough to "expel Constantine's magistrates." 

Expelling a magistrate. This reminds me of current events!

But, seriously, go pound sand. 

Amongst all the 200+ explanations of the end of the Roman Empire, it is pretty hard to get around the whole "end of the standing army" thing. Right back at the beginning, for example, it was Edward Gibbons' explanation. Here's the direct quotation, direct from the linked Wikipedia article:

The story of its ruin is simple and obvious; and, instead of inquiring why the Roman empire was destroyed, we should rather be surprised that it had subsisted so long. The victorious legions, who, in distant wars, acquired the vices of strangers and mercenaries, first oppressed the freedom of the republic, and afterwards violated the majesty of the purple. The emperors, anxious for their personal safety and the public peace, were reduced to the base expedient of corrupting the discipline which rendered them alike formidable to their sovereign and to the enemy; the vigour of the military government was relaxed, and finally dissolved, by the partial institutions of Constantine; and the Roman world was overwhelmed by a deluge of Barbarians.

Gibbon uses two loaded terms here, "partial institutions of Constantine," and "discipline," which together slip in an argument that the Christian religion was bad for the Roman Empire, for which &etc.

Curiously, what Gibbon doesn't talk about is subsistence and tax regimes. Gibbon is a lot closer to the age in which 90% of the population worked on the land. He has absolutely foregrounded army pay, and, while he does not give give us a picture of Roman Britain, because that is not his working method, he absolutely does describe the "state of Germany," which was, as far as he can reconstruct, primitive. Germany has few cities compared with the late Eighteenth Century, rather living in "rude fortifications raised in the centre of the woods," and hut, raising cattle and "scant quantities" of grain, wearing furs and knowing nothing of orchards or water meadows. Britain, and even Roman Britain, should be broadly comparable. It's not quite Conrad's vision of the Roman Thames as the Nineteenth Century Congo, but good enough for grocery work. 

The pendulum has swung very far in a different direction nowadays. (At least allowing that I am putting words in Gibbon's mouth.) No longer a land of primeval forest, archaeologists envision an ancient, indeed, Neolithic Britain as one with roughly the same amount of arable as today. Population is hard to put one's finger on when rounding errors are bigger than numbers, but Francis Pryor's Britain AD has an eye-catching binding, an index, and someone to disagree with, so I will put in his numbers: between 2 and 2.5 million Britons at the beginning of the Roman Conquest, up to 4 million at its end, with Pryor disagreeing with Heinrich Haerke's theory that this 4 million declined to 2 during the Late Roman period.

I am having trouble imagining how Gibbon would have reacted to this. The "90%" figure would be problematic. The obvious question, for Honorius and for that "former lieutenant of the Norfolk militia" alike is: why couldn't the cities of Britain raise their own standing army?  Gibbon actually answers this, for what it's worth: It's Constantine's fault. He knows what happened, even if he might be selecting the wrong scapegoat.

We don't. Basically, because we've lost touch with the facts on the ground.

Areas of in Various Crops of the United Kingdom in 1875 (millions of acres) (NB: area of the United Kingdom and Republic of Ireland in acres: 59,700,000 acres).

Crop Acreage	Wheat	Barley	Oats	Rye	Legumes	“Green Crops”1	Grass(2)	Grass(3)
Britain	3.5	2.8	4.2	100,000?	0.8	5.0	6.0	23.7

1.        Potatoes, turnips and swedes, mangel, cabbage and other cruciferous, vetches and tares, “other green crops.”

2.        Grass broken up by rotation

3.        “Permanent” grass, excluding grazed heather and moor (another 11 million acres.)

These numbers aren't completely transparent, because Nineteenth Century British farmers practice "crop rotation."

The Norfolk four-field rotation

Season	Year 1 (5, 9. .), Fall	Summer, Year 2	Fall, Year 2	Winter, Year 4	Summer, Year 4
Crop Removed/Harvested	Clover ploughed up	Wheat harvested	Catch crop, if any	Clover and barley	Barley harvested
Crop Planted	Wheat	“Catch crop,” ie. rye, vetches, winter oats, buckwheat, etc. on light soils only	Root crop		Clover left in ground

The Rothamstead Experimental Station determined that an average English field, kept continuously in wheat, should produce the average yield of 28 bushels of 60lbs each for eight years, then begin to decline at one-half bushel per acre per year. A course of a  few seasons of "green manure," in crops such as mustard, or peas and beans, would restore its fertility, and, for reasons to be discussed, it would not much hurt to use these fallow crops for feeding animals once they were mature.

The four-course roation works very differently, but has the same object in view. The discovery that farm fertility was dependent on a hidden chemical economy of phosphorus, potassium and nitrates puts the traditional wisdom of not exporting the fertility of the farm on a scientific basis. Growing cereal for sale removes ions from the soil. Recycling crops into fodder, and selling meat,dairy and wool drastically reduces this export, allowing the farmer to maintain and even increase soil fertility, and maximise cereal production in the long run. This incidentally leads to the production of a great deal of fodder, and so of livestock, just as does the ancient farm-until-exhausted-then-shift. But is that really such a bad thing?

Yes, of course it's a bad thing. Raising meat is always wrong (I'm not a vegetarian, I just like telling you what you can eat); Only the reasons change.

I am also going to point to the catch crop-root-crop-barley sequence shows the importance of soil conditions and ploughing technique. Light soils in southern climates will take the additional crop. On heavy soils, like the east Anglian clays that inspired the Norfolk rotation to begin with, the farmer needs all the time available to finish plowing prior to seeding the root crop. Barley follows on the root crops (which, remember, whatever their benefits as animal fodder, are planted because it is possible to harrow between the rows for weed control), because barley is a shallow-rooting crop which does not require that the soil be turned over for planting, a shallow "stirring" being sufficient. 

This is important because there was a thing in the old days about the replacement of the "primitive" ard with the more "advanced" turnplough.

By Anand S - https://www.flickr.com/photos/anand_bdr/16918825146/, CC BY-SA 2.0, https://commons.wikimedia.org/w/index.php?curid=39238064

Scare quotes aside, the turnplough does require a honking big piece of iron. The discovery of iron tools (pdf)

Scraped from an online publication of James Curle, A Roman Frontier Post and its People (Macklehose and Sons for the  Society of Antiquaries of Scotland, 1911), linked above.

 including ploughshares in hoards in late Empire "refuge fortifications" in southwestern Germany inspired Nineteenth and early Twentieth Century agricultural historians, when they were not explaining the whole of French social history, to see iron ploughshares moving from Germany to England. The whole "Anglo-Saxon" invasion might be imagined as Germanic immigrants spreading up the river valleys while Roman British looked down from their light-soiled farms along the spring line!

As I say, this is an old argument, and tt took me a long time to find a discussion of it (for the same,revised with 100% more Celticness, see here). When I did, I was surprised by a very recent and robust statement. It even features invading Germanic hordes burning the motherfucker down, an aspect that has gone out of fashion almost as much as the idea that replacing ards with turnploughs represents unequivocal technological progress. I'll leave Janine Freis-Knoblach and Heiko Steuer to defend their argument over at Google Books, but I am impressed by the very local character of the argument, in which the German invaders are sometimes envisioned as climbing the hill from their riverbottom farms to burn out their ard-using Roman villa-living neighbours. Plough technology replacement as social revolution! Maybe Marc Bloch was right?

Right or wrong, the distribution of Roman roads and Roman farms is what it is. This is not a story about "progress," except insofar as iron making is very definitely technological progress. What we see here is prioritisation of the best land for cultivation under a given labour management regime. We just have to figure out what that regime was. (And, given the week's news, get some joy out of the idea of this hazily sketched "populist revolt.")

Let's look at typical Nineteenth Century crops:

Yield of major British crops in lbs/acre. 

Crop	Weight of Crop
	Wet		Dry
Wheat (30 bushels) Wheat straw	1800		1530
	3,158		2653
Barley (40 bushels) Straw	2080		1747
	2447		2080
Oats (45 bushels) Straw	1890		1625
	2835		2353
Maize (30 bushels) Straw	1680		1500
	2208		1877
Meadow Hay, 1 ½ ton**	3360		2822
Red Clover Hay, 2 tons	4480		3763
Beans, grain, 30 bushels, Straw	1920		1613
	2240		1848
Turnips, roots, 17 tons, Leaf	38,080		3349
	11,424		1531
Swedes, roots, 22 tons, Leaf	31,360		3349
	4,704		700
Mangels, root, 22 tons, Leaf*	49,280		5914
	4,704		1654
Potatoes	13,440		3360

*Our author notes that crops of 100 tons mangels/acre must be treated as exceptional. 

**There are two mows a year in the British climate.

Food Value of various foods

Digestible Matter in 1000lbs of various foods

Food	Total Organic Matter	Nitrogenous Substances		Fats	Soluble Carbohydrates	Fibre
		Albminoids	Amides, etc
Cottonseeed cake (decorticated)	691	374	18	128	158	13
Linseed cake*	422	150	13	50	177	32
Peas	655	230	11	103	266	45
Beans	747	196	25	12	499	36
Wheat	733	196	28	12	446	51
Oats	600	81	7	45	441	26
Barley	715	70	4	19	607	15
Maize	715	70	4	19	607	15
Rice meal	612	67	10	44	651	12
Wheat bran	585	90	20	27	426	22
Malt sprouts	681	114	71	11	379	106
Brewers grains	137	34	2	14	67	20
”” dried	529	136	8	57	266	62
Pasture grass	156	19	11	6	84	36
Clover (bloom beginning)	123	17	8	5	63	30
Clover hay (medium)	440	47	25	13	242	113
Meadow hay (best)**	511	60	18	13	269	151
“”” (Medium)	485	40	18	13	269	151
“”” (Poor)	460	29	5	10	242	174
Maize silage	124	1	7	7	75	34
Bean straw	412		40	6	211	155
Oat straw	381	7	5	7	163	155
Barley straw	426	4	3	6	211	202
Wheat straw	351		4	4	150	193
Potatoes	213	5	9	1	195	3
Mangels (large)	89	1	8	½	74	6
“”” (small)	109	2	6	½	96	5
Swedes	87	2	7	1	71	6
Turnips	68	1	5	1	56	5

*It's fun to call "flax" by another name, and, God knows, we need agronomy to be more confusing.

**Although we tend to think of hay as a less productive crop than wheat, it has wheat beat all hollow as a food --for digestive systems which can support it. 

 "Mangelwurzel." Closely related to sugar beets.  Wiki. User:MarkusHagenlocher - Own work, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=1089069

Number of Horses, Cattle, Sheep and Pigs in the United Kingdom

Year	Horses*	Cattle	Sheep	Pigs
1875	1,820,000	10,163,000	33,492,000	3,495,000

 *For some reason that makes sense to the author, “horses” includes only unbroken, brood mares, and “horses exclusively used in agriculture”

Number of Horses, Cattle, Sheep and Pigs in the United Kingdom in 1905, broken down by principality

England	1,204,000	5,021,000	14,698,000	2,083,000
Wales	161,000	739,000	3,435,000	211,000
Scotland	206,000	1,227,000	25,527,000	130,000
Ireland	535,000	4,645,000	3,749,000	1,164,000

The first thing that would make Mr. Gibbon clear his throat and make his country squire's contribution to the discussion is the notion that 2.5 million (or 2 million, or 5 million) ancient Britons were tending 11.9 million acres of arable crops and 30 million acres of pasture, and, under "modern" conditions, they would also be responsible for 40 million head of livestock (poultry and sundry excluded.)

Also, they would be producing 7.5 million tons of cereal by a very rough estimate, using, recall, estimates of the traditional, pre-chemical fertiliser, pre-chemical pesticides crop yields. Before the "catch crop."

Needless to say, this is more work than even a "modern" agricultural labour force could supply.

Number of “Persons engaged in Agriculture,” in the United Kingdom, 1851—1901.

1851	1861	1871	1881	1891	1901
3,453,500	3,080,500	2,744,000	2,573,900	2,394,500	2,262,600

The default is to assume that in a less efficient, Roman-era agricultural economy, there was less arable and more livestock, but the decline in the British agricultural labour force in the last half of the Nineteenth Century was driven by the decline in arable (from 11,399.030 in 1875 to 8,333,770 in 1905, for example. The British Census for 1851 is available online but does not distinguish arable from pasture in its total return of 24 million acres under cultivation.)

 So what do we do with the suggestion that it was all cleared and incorporated into field systems? Even if we assume that it was "actually" only a part of England only,

Provenance here is a little strained. I lifted it from a Reddit post by igreatplan because that was easier than doing a photographic reproduction of the same figure in Pryor's Britain AD, but I doubt that it is original to Pryor, either.

we are still talking about a very large proportion of the reported arable. (I'm being weaselly because I could work it out, and choose not to do so). There is no way that a population of 2--5 million Roman Britons could have cultivated this much land. No way.

If 90% of the Romano-British population was on the land, by the figures presented here, 90% of the Romano-British population was fucking the dog.

Which, actually, we knew. Or knew at one point. Mr. Gibbon takes it for granted that Iron Age and Roman farmers were inefficient. He should know! His whole period was one of "improvement," striving to bring the agricultural interior of his native island into the market, increasing production, rents, and improving the living conditions of the poor. Roads transform the poor rural districts of Scotland, for example, by allowing wheeled vehicles to reach the farm, allowing it to produce bulky crops for sale on the market.

It's only in more recent eras, bewitched by the programmatic theories of the Reverend Malthus, according to which human populations instantly expand to the "carrying capacity" of the landscape to maximise human immiseration, that we have led ourselves into this ridiculous paradox in which a population which is 90% involved in primary production somehow cannot produce the tax revenue needed to support a standing army.

I mean, if 90% of the population is unemployed --that's one thing! (A thing soon to be tested in a modern industrial economy near you!) If they're working, however. . .

Yes, yes, I know: the immiseration theory holds that the population has expanded to the point where the labour force consumes what is produced. Two comments on this: first, and most trivial, the analysis above suggests that the "subsistence economy" of Great Britain is not what we assume it to be --specifically, it will involve somebody eating meat, and that somebody will have to be peasants, heresy of heresies!--. More importantly, it is not a story compatible with a "subsistence" population of less than 10 millions or so; and the higher we push the population in our search for the Malthusian nightmare, the higher our tax revenues must rise, however small be the surplus to be extracted, and the greater the number of legions to be employed.

The last thing this primitive economy lacks  is the taxable produce needed to support an army. So it follows that Roman taxes are low, even though they were, in fact, a heavy burden on the population. Faced with this paradox, Gibbon robustly assumed the obvious: that taxes were low in terms of the revenue extracted, but high in proportion to their impact on the population.  ("Partially reformed institutions of Constantine.") This does not mean that the taxes took a high proportion of the potential output of the land, as laid out in all those tables above. On the contrary, Low taxes may have caused low agricultural productivity --or surplus production, anyway. In which case, a defective tax regime may make state revenues, and peasant living conditions, much lower than they could be.

So Honorius has a  pretty good point, actually. The point worth further exploration here, is how ancient Roman economic institutions came to be so structured as to be such a significant drag on production.