A favourable evolution of the climate seems to have played a role here, ³ but technological advances may have been far more important. Ploughs were improved in the second half of the first millennium, which was the more necessary as the soils of northern Europe are heavier than those of the Mediterranean basin. Wheeled ploughs made it possible to regulate the depth of the furrows, plough shares made of iron or covered with iron made for greater efficiency and durability of the equipment. Mould boards served to turn the soil opened up by the plough shares. The type of yoke used in the pre‐christian Mediterranean world was suitable for oxen but less so for horses, since if the load was heavy it asphyxiated them. It was only the introduction of the horse collar that permitted horses to be used for ploughing, for which they are better suited than oxen, and for pulling heavy carts. (The horses known to the pre‐christian pólis‐world—and, presumably, the Roman world, too—were considerably smaller than the breeds to which we are used nowadays, and ill‐suited for ploughing and heavy loads in any case; conversely, to plough the lighter soils of, e.g. Attikế, mules could be used as well as oxen.) ⁴ No more than the horse collar did the pre‐christian Mediterranean world have horse shoes or stirrups. They arrived in western Europe in the ninth or tenth century, having originated, like the horse collar, in Asia. Stirrups allowed the rider to handle his weapons more effectively, and also enabled him to wear heavier armour, for which the breeds of horses by then available were large and strong enough. More importantly, from an economic point of view, horse shoes further increased the efficiency of horses as draught and pack animals. ⁵

But a still more significant and basic factor contributing to the economic and demographic dynamism of Latin christendom and indeed ancien régime Europe right up to the onset of industrialization was the spread of watermills. The old Mediterranean civilization was run on very little energy. By the end of the ancien régime, Europe had far more energy available to it than did the Graeco‐Roman world; the reason was mills. Though we are familiar with the enormous impact on society of the increased availability of machine power resulting from the invention of the steam engine, the fundamental importance of the history of milling for the evolution of western civilization, including its political structures, is far from having received the attention that it deserves.

The kind of grain that, from its inception in the Greek world, has been the main nutritional basis of western civilization cannot be eaten raw, nor can it be cooked (or baked) as it is. It has to be ground first, for which mills are needed. In the pre‐christian Mediterranean world, for a long time milling was done by hand. A significant portion (p. 343 ) of the available labour had to be invested in milling. ‘[F]ifty slave women he had in the house, of whom some grind the yellow grain on the millstone, and others weave fabrics, or, as they sit, twirl the yarn’: in describing the residence of the king of the Phaiaks, a very grand lord with whom Odysseús is staying at one point in the course of his travels; Homer in the eighth century bce lists milling as the first of two major tasks of the slave women, before the production of textiles. ⁶ The royal residence of Odysseús himself, on the island of Ithákê, is somewhat less grand, requiring less flour: at the mills there ‘twelve women in all plied their tasks, making meal of barley and of wheat, the marrow of men’. When Odysseús, unrecognized, returns after a twenty‐year absence, he overhears one of those women curse her bone‐breaking work late at night—being the weakest, she is still busy with it even as the others have gone to sleep. She interprets a distant thunder as an omen from Zeus that the days of the suitors, who have taken over the royal residence while they are trying to get queen Pênelópê to marry one of them, are numbered. ⁷

For this woman, reducing the number of mouths to feed was the only way to lessen her workload. It was not until many generations later that the harnessing of water power and wind power would set free an enormous amount of labour that, like hers, hitherto had unavoidably been tied up in milling. Moreover, once this technology had been established, it proved the catalyst for many further improvements, liberating still more labour while at the same time increasing productivity in every production process to which it was applied. Slow to get under way in the Graeco‐Roman world, in the period of Latin christendom this development eventually gained enormous momentum.

4.1.1.1 Tentative beginnings: watermills in the Greek and Roman world

The mills operated by the women mentioned by Homer had an upper millstone or runner stone that was moved back and forth across the lower millstone, the grindstone or bedstone (sometimes spelled ‘bidstone’). Such mills consumed a lot of muscle power, since the runner stone had to be continually stopped and re‐accelerated, a great waste of momentum. Even with the levered version that appeared some time after Homer, the back‐and‐forth movement of the runner stone remained intrinsically wasteful of energy. The invention of revolving mills—operated by hand, or by draught animals—was thus a breakthrough in itself. Here, the milling could be carried out in one continuous movement. Not introduced until the fourth century bce, revolving mills were also the prerequisite for the invention of watermills, which appeared in the third century bce at the earliest.

Millstones have to be horizontal. In the simplest type of watermill—the turbine mill—the mill wheel, built like a propeller or ship's screw, is horizontal, too. Linked to the revolving runner stone by a vertical axle, it requires no gearing. More refined than the kind of turbine mill common in parts of Europe in the more recent past, the Roman mill complex of Simitthus (Chemtou in what is now Tunisia) consisted of three circular shafts of masonry that ensured that no water could bypass the horizontal mill wheel at the bottom, while also acting as pressure tanks. Water was fed into the shafts tangentially, giving it a circular movement from the start and, because the ducts were tapered, accelerating it in the manner of a jet. The water came from the Bagrada (p. 344 ) river (now Oued Medjerda), which was dammed at this point by a collapsed bridge that archaeologists think was last restored in the late third century. Evidently the mill complex was built later, yet seems to have been abandoned as early as the first half of the fourth century. A nearly identical complex at Tichilla (now Testour) on the same river has been dated to the same period. However, no further example of this type of mill has yet been discovered anywhere, either from the Roman period or of later date, suggesting that it was the work of a local inventor that failed to spread beyond its area of origin. If it had spread, the technology, as efficient as it is simple even as it seems to bear the signature of an experienced engineer, would no doubt have continued to be employed after the demise of the empire, so that recent examples of the same type would have been found. ⁸

The type of mill wheel most familiar from Europe and North America is vertical (i.e. the axle is horizontal). This requires gearing to transform the vertical rotation of the waterwheel into the horizontal rotation required for milling. The two basic varieties of the vertical mill wheel are the undershot wheel and the far more powerful overshot wheel. In the case of an undershot wheel, the water hits the wheel paddles in the lower section of the wheel, which can simply be hung into the current. It thus requires little site preparation, least of all in the case of the ship mill, which has an undershot wheel mounted between two pontoons anchored in a river and which in much of Europe was a familiar sight well into the nineteenth century. The water for running a mill can also be supplied by a mill race, indispensable in the case of the overshot wheel, which the water hits from above. If the mill race makes the construction of the mill more complicated, in return it makes it possible to control the water flow (by means of a weir), and to tap even quite small streams. In the case of the overshot wheel, a big additional advantage is its high power output. ⁹

It is generally taken for granted that the overshot wheel was known in the Roman period, but the evidence to me seems to point the other way. In Capital (Chapter 15 Section 3b), Karl Marx cites a poem ¹⁰ by Antípatros of Thessaloníkê dating from the latter part of the first century bce and which Marx presents as an early example of machine power being regarded (wrongly, in his view) as enabling people to work less. He quotes the poem in the 1782 German translation of Christian zu Stolberg, rendered as follows in the 1887 English edition of Capital edited by Friedrich Engels:

• Spare the hand that grinds the corn, oh, miller girls, and softly sleep.
• Let Chanticleer announce the morn in vain!
• Deo has commanded the work of the girls to be done by the Nymphs,
• and now they skip lightly over the wheels,
• so that the shaken axles revolve with their spokes
• and pull round the load of the revolving stones.
(p. 345 )
• Let us live the life of our fathers, and let us rest from work
• and enjoy the gifts that the Goddess sends us.

‘Deo’—Dêố;—is the Homeric name for Dêmếter, goddess of grain, but also in charge of the nymphs, who in Greek mythology were water spirits associated with sources, streams, and the like. The Greek text describes the nymphs as kat'akrotátên halloménai trochiến, which may be rendered as ‘jumping (or skipping) along the topmost (or outermost) (part of the) wheel’. Andrew Wilson, ¹¹ in claiming that the phrase means ‘jumping down onto the topmost part of the wheel’, and that the nymphs are simply identical with the water, overlooks the alternative reading adopted by Stolberg. This has the nymphs 'skipping over’, or dancing on the wheel, as if rejoicing in the power of the water, from which they are distinct, but with which they are associated. That seems more in keeping with the traditional image of nymphs, and is applicable to either type of vertical mill wheel: even if the water flows underneath the wheel the nymphs can still frolic on top of it. The poem is thus entirely compatible with the assumption that the overshot wheel was not yet known, and no proof at all that it was.

Wilson further points out that Antípatros is alluding to a geared mill, and that this rules out a horizontal mill wheel since that does not require gearing. In fact, it does not follow that such a mill cannot be geared; archaeological evidence of a second‐century geared horizontal mill will be cited in a moment. In Antípatros’ poem, it is not the fact of gearing that rules out a horizontal mill wheel but its kind. The reference to gearing consists in the words aktínessin heliktaís, a dative of instrument which literally means something like ‘by means of rotating (or: swirling) rays’. Wilson believes this to be a reference to ‘ “encircling” or “rotating” cogs’. That seems to be stretching the meaning of the word ‘rays’. Stolberg, more correctly I think, has ‘the shaken axles revolve with their spokes’, where the words in italics correspond to aktínessin heliktaís. He may have got the image of spokes from the ‘rays’—something radiating outwards from a centre, as spokes of course do: indeed radius is the Latin for 'spoke’—being described by Antípatros as ‘rotating’ or 'swirling’, like for example the rotating spokes of the carriage wheels with which Stolberg (1748–1821) would have been familiar.

But where are the spokes in a mill? In the mill wheel? Perhaps Stolberg thought so, but there is another solution. In the 1940s, archaeologists discovered the remains of a watermill belonging to a second‐century Roman villa (a country estate) near Hagendorn in the Swiss canton of Zug. Part of the finding was at first identified as wagon wheels: wooden rods looking very much like spokes radiating from axles. On closer examination the spokes seemed rather short for wagon wheels. There was no trace of a rim, and the axles turned out to have been connected to a waterwheel and a revolving millstone, respectively. The contraption was identified as what the archaeologists’ report (not published until 1991) ¹² calls a Sterngetriebe or star gearing: the axles were set at right angles, geared together not by means of cogs but by means of the pointed 'spokes’ inserted into (and thus radiating from) sturdy discs around the axles. As the spokes were quite numerous, even a relatively slow rotation would produce the 'swirling’ effect that Antípatros may be alluding to. No one seems as yet to have established a connection between Antípatros’ poem and the Hagendorn find; but I suspect that he had a star gearing in mind.

(p. 346 ) With a star gearing, at least of the Hagendorn type, both axles turn at the same speed. It is apparently still found, or was found until recently, in the Near and Middle East—not for milling, but for pumping water: think of a vertical waterwheel, with buckets on the rim scooping water, and driven by a donkey turning a horizontal wheel. Here, the fact that the waterwheel turns no faster than the donkey walks presents no great disadvantage, there being no need for the waterwheel to rotate particularly fast. In milling, however, the faster the rotation of the millstone the better. Clearly, the star gearing documented at Hagendorn and, with some likelihood, described in Antípatros’ poem, represents an early stage of watermill technology. This is in keeping with the fact that Antípatros adopts the pretence of telling the ‘miller girls’ something new. Of course, real mill operators would not have needed a poet to advise them of this new invention. At the same time, however, if the invention had been around already for generations, the poetic pretence would not have worked.

In interpreting Antípatros’ ‘rays’ as ‘cogs’, Wilson was no doubt influenced by a famous passage in Vitruvius (10.5.2), which describes a gear box for a mill driven by an undershot waterwheel. Vitruvius unambiguously speaks of ‘toothed discs [tympana dentata]’, of which the one on the axle driven by the waterwheel was the larger (maius): in other words, the millstone rotated faster than the waterwheel. It is striking that Vitruvius was familiar with this kind of gearing, much more suitable for milling than the star gearing, but Antípatros writing more or less at the same time apparently was not. Vitruvius was an expert engineer. Antípatros, on the other hand, must have based himself on what he was seeing around him, which suggests that the relatively few watermills that he was able to inspect (they must have been relatively few or he could not have presented them as a novelty) typically had the more primitive star gearing.

Both Vitruvius and Antípatros were active in the second half of the first century bce. Wood from the Hagendorn mill has been dated with the help of the tree rings: it was cut in the year 176 ce. So even then news of the superior gearing described by Vitruvius had not reached the Hagendorn villa. The reason for this was not its geographical remoteness from the core areas of the Graeco‐Roman world, since evidence of more advanced technology from the same period has been found still further north. In 1912, archaeologists discovered the remains of a Roman mill at Zugmantel near Frankfurt‐am‐Main, on the very edge of the Roman empire. Situated on a hilltop and associated with a Roman fort, this mill was not water‐powered but turned by humans or animals. Yet it possessed a whim: a gearing with wooden gear wheels of different size but parallel axles. In other words, the sole purpose of this gearing was to speed up the rotation of the millstone, not—as is necessary in a watermill—to convert the vertical rotation of the waterwheel into the horizontal rotation required for milling. In 1996, another Roman mill was found near Aschheim in Bavaria. Consisting of a wooden frame, about a metre high, that supported the millstones, it had a metal crank turned vertically by hand and a gearing that, if the archaeologists have reconstructed it correctly, already looked exactly like the standard mill gearing of the late pre‐industrial age. A large cog turned by the crank was geared to a small pinion on the axle of the millstone, which therefore rotated much faster than the crank. This pinion—like the Zugmantel one—was not toothed, but of the ‘lantern’ type, a trundle: two parallel discs connected by sturdy cylindrical bars along their rim. The large cog of the Aschheim mill likewise was not toothed along the rim but, like the standard large cog wheel employed by the millwrights of the ancien régime, consisted of (p. 347 ) short, thick pegs set vertically into the rim on one side of the wheel but without sticking out beyond its circumference.

This gearing is not the one described by Vitruvius. The problem with toothed cogs (tympana dentata) would have been to keep them geared precisely to each other, since the pressure brought to bear on them by the inertia of the waterwheel and of the millstone, and the torque of their axles, would easily either shift them out of position or cause damage to the teeth along the rim—a late ancien régime gear wheel for a mill would typically have replaceable teeth and pegs made from softer wood (such as fruit wood) than the wheel frame so that in the event of some mishap they would shear off without damaging the frame. The shorter the teeth, the more robust the cogs. Short teeth required a sturdy construction ensuring that the cogs were held in position securely, since even a little play would incapacitate the machine. Long teeth allowed for more play but were more easily breakable. A star gearing (which as it were consists of very long teeth) allowed for maximum play but would have been least pressure‐resistant. Clearly, the abandoning of protruding teeth in the gearing of the Aschheim mill is the fruit of some experience with milling and of adaptation to the stresses to which the machinery is exposed—and that despite the fact that, as I think is likely, the overshot wheel with its high energy output had not even been invented. The torque exerted by an overshot wheel is notoriously high; probably too high for a star gearing. The ‘circumference gearing’ represented by the Aschheim mill is less affected by torque. Used even for a handmill as in the Aschheim find, it is no indication that the overshot wheel was known—though it was a precondition for its eventual introduction. ¹³

Both the Zugmantel mill and the Aschheim mill are dated by the archaeologists to the second century, which means that they are roughly coeval with the Hagendorn mill. The coexistence of the advanced gearing of the former two with the primitive star gearing of the latter to my mind suggests that advanced milling technology, which all three of them display (the Hagendorn mill by virtue of its power source), was rare, preventing the emergence of a common technological standard. Perhaps this also explains why (apparently) innovative technology such as that found on the Bagrada river failed to catch on and spread to other areas.

When the Goths besieged Rome in 537, they cut the aqueducts serving the city. This did not lead to any shortage of drinking water, only to the closure of the public baths. More inconveniently, it caused a shortage of flour: served by one of those aqueducts, the Aqua Traiana, the watermills on the Ianiculum hill (now Gianicolo) were left without power, forcing the defenders to build ship mills on the Tiber. The historian Prokópios, in his eyewitness report of that war, speaks of the Ianiculum as ‘a great hill where all the mills of the city have been built from old, because much water is brought by an aqueduct to the crest of the hill, and rushes thence down the incline with great force’. ¹⁴ Still in good condition, the Traiana now supplies a monumental fountain on the slope of the Gianicolo built in 1605 by pope Paul V. Since the aqueduct brings water to a hillside, there is so much height of fall available here that any late ancien régime engineer commissioned to use this water for driving mills would as a matter of course have equipped them with overshot wheels, which require a lot of head (the difference in height between the (p. 348 ) ‘headwater’ upstream of the mill and the ‘tailwater’ downstream) but yield far more energy than undershot wheels. A third‐century complex of mills drawing on the water of the Traiana and partly excavated in 1998 is the only evidence of the Roman mills of the Gianicolo so far discovered. It was equipped only with undershot wheels. ¹⁵

Overshot mill wheels are generally assumed to have turned in the great mill complex of Barbégal near Arles in southern France, variously dated to the second, third, or fourth century. An aqueduct brought water to a ridge overlooking a plain. As the water leapt downhill over eight terraces set into the hillside, it apparently powered twin mill wheels, separated by a stairway, on each of those terraces. Almost nothing now remains of the architecture of the complex, so not only is it not clear how the machinery was housed, but also there is nothing to indicate what type of mill was employed here. If the many reconstruction attempts unanimously posit overshot wheels, it is because their authors assume the Roman builders of the complex to have realized that overshot wheels are the most energy‐efficient and that the height of fall provided by the hillside practically cries out for them. But on closer examination that assumption seems questionable.

The total height of fall available here is 18.6 m, or 2.33 m for each of the eight wheels: for overshot wheels, that is just about feasible, but suboptimal. 2.33 m corresponds pretty much exactly to the minimum head of four ‘ells’ (Ellen) posited for overshot wheels in the self‐consciously scientific handbook on milling technology of Leupold and Beier of 1735—anything less, the handbook explains, and the wheels would have to be very broad. More recent literature tends to give a slightly higher minimum (such as 2.5 m). But overshot wheels are the more powerful the larger they are—where enough height of fall was available, in the late ancien régime they routinely had a diameter between 5 and 10 m. Anyone wanting to operate overshot wheels on this site and who knew what they were doing would have built seven terraces at most, and almost certainly fewer. That of course would have meant fewer machines, too—but more powerful machines, with less construction and maintenance effort: thus, if the mills were gristmills (as fragments of millstones on the site indicate), with larger overshot wheels their throughput of grain would have been higher.

Now what about simply positing undershot wheels (or ‘breast wheels’, which the water hits halfway up their total height, and which the archaeologists’ report on the Hagendorn mill speculates may have operated both there and at Barbégal)? All difficulties disappear: 2.33 m of head is ample for them, nor would there have been a premium, in this case, on having larger rather than more wheels: larger undershot wheels would have made sense in a big river with its unlimited supply of feeding water, but not given a relatively sparse supply as here (overshot wheels do not have this problem as they make better use of limited feeding water). Even though, at Barbégal, overshot wheels would have been physically possible, they would indicate a lack of expertise concerning them. The set‐up with its eight terraces more plausibly suggests a scenario in which whoever planned this site was, as a matter of course, thinking in terms of undershot wheels (or perhaps breast wheels)—quite likely because overshot wheels were as yet unknown. ¹⁶

(p. 349 ) 4.1.1.2 Takeoff: water and wind power as the motor of post‐Roman western civilization

What drives overshot waterwheels is almost exclusively the weight of the water filling the cell‐like compartments or ‘buckets’ along their rim and pulled down by the force of gravity, rather than the impetus of the current captured by the paddles of undershot wheels: in the jargon of physics, undershot wheels are powered by the kinetic energy of the water, and overshot wheels by its potential energy. That overshot wheels are the more efficient type is not intuitively obvious—especially as they turn slowly, as if lazily. To understand why they are more powerful one has to remember that the impetus provided by flowing water is likewise solely due to the force of gravity (for doing which of course it is necessary to have the concept of gravity in the first place, as, if I am not mistaken, the Graeco‐Roman world did not).

Unless we are talking about a waterfall, a natural watercourse like a stream or river has only a moderate to imperceptible incline. Whereas in a horizontal waterwheel the Coriolis force may play a minor role depending on its construction, what drives a vertical waterwheel, of any construction, is ultimately the force of gravity and nothing else; and that force is the greater the greater the incline along the distance over which the water acts on the waterwheel. With this in mind, it becomes clear why, given the same quantity of water acting on the wheel per unit of time, the overshot wheel makes more efficient use of it than the undershot wheel. The vertical distance travelled by the water while it pushes the overshot wheel is far greater. Moreover, trapped by the ‘buckets’ of the wheel, the water cannot escape from acting on the wheel, whereas part of it does just that in the case of the average undershot wheel (unless there is a mill race and the wheel is encased to keep all the water acting on the paddles). The energy output of a waterwheel is the product of the weight of the water acting on the wheel per unit of time and the vertical distance that the water travels while acting on the wheel. Other things being equal, an undershot wheel turns faster than an overshot one, which means that during the same period it ‘processes’ more water than does the overshot wheel. This in part compensates the fact that it operates with a far lesser height of fall. Nevertheless, the energy output of the overshot wheel is several times (five or more) that of a comparable undershot wheel.

The overshot wheel is subject to greater constraints than the undershot one, which means that not infrequently only an undershot wheel is possible—starting with the necessary minimum head that in flat terrain is usually not available. Further, the slow rotation of the waterwheel necessitates a gearing with a high speed ratio to ensure that the upper millstone rotates fast enough. Moreover, less incline in a watercourse often means more water—a torrent in the mountains carries far less water than a mighty river in the plain. If the discharge (quantity per unit of time) and flow rate of the feeding water is not a problem, the low energy output of an undershot wheel can be compensated by simply placing a number of them one after the other, since each ‘uses up’ so little head. Alternatively, in the case of a ship mill, or a mill built between the pillars of a bridge as also was often done in Latin christendom, it is possible to equip it with a very broad wheel, or again use several, set side by side.

In the famous passage cited earlier, Vitruvius explains that the mills discussed by him were powered in the same way as the bucket wheel (for irrigation) that he describes immediately before; this was, and to some extent still is, common in the Near and Middle East. Known as noria in Arabic, it is basically a paddle wheel driven by the current—that (p. 350 ) is, with respect to the way it is powered it is an undershot wheel. Since mill builders were initially interested in a power source, a source of rotational kinetic energy for milling, and since it is the paddles, not the buckets, that cause a bucket wheel to turn, it is easy to see why they happily adopted the paddle wheel for milling and did not bother to reflect whether an even better, if less obvious, power source might not, as it were, be hidden in the buckets, which instead were got rid of. (The breast wheels that may have operated at Hagendorn or Barbégal would have had paddles like undershot wheels, not buckets like overshot wheels.) The buckets of a noria empty the water scooped out of the river that makes the wheel turn into a conduit connected to an irrigation system. It seems to have required more lateral thinking than anybody could be bothered to invest in to grasp that instead of powering the wheel—by means of paddles pushed by the current—to empty the buckets it was possible to do the reverse: fill the buckets to power the wheel. This leap of the imagination was the less likely as the kind of river powering a noria—like the Orontes (in Arabic, Asi), where they are still quite numerous—did not permit overshot wheels owing to its slight incline. Most likely, the requisite mental breakthrough came only once watermills had become ubiquitous, which in the Roman world was not yet the case, and were routinely experimented with by a considerable number of people specializing in their construction.

Watermill technology was less simple than in retrospect it may appear. By the standards of the pre‐industrial age, the overshot mill, in particular, set free a vast amount of kinetic energy, which could cause havoc if not controlled properly. We have already discussed the need for sturdy gearing. By the late ancien régime, a good watermill would have a large ‘pit wheel’—so‐called because it was often partly sunk into the floor of the mill house—mounted on the same axle as the waterwheel. The pit wheel was geared to a small pinion and thereby drove a horizontal ‘great spur wheel’, a large, fast‐revolving toothed wheel on the floor above and mounted on the same vertical axle as the pinion. The spur wheel in turn meshed into the 'stone nut’, another small pinion driving the runner stone (or several, each with its own pinion and set along the circumference of the spur wheel). The double transmission ensured that the runner stone rotated twenty‐five to thirty times faster than the waterwheel: for milling flour with horizontal millstones a minimum of some fifty revolutions per minute is apparently necessary, but a good mill in the late pre‐industrial era would achieve 120 or so. That required sturdy metal bearings lest some fast‐spinning axle pierced its worn‐out bearings and set the wooden machinery on fire. Nor is a millstone indestructible: the stresses on the stone rise as the square of its rotational speed. There was a considerable risk of its disintegration, and a lump of stone with an average weight, in the late ancien régime, of two to three tons breaking up as it turns at high speed is not an experience one would want to repeat. Moreover, a high‐quality millstone—often imported from far away, and made up of several pieces joined together—could cost a small fortune. It was commonly secured with iron hoops around its circumference. ¹⁷

The higher the rotational speed of the runner stone, the more important was it that it never touch the bedstone, even though at the same time the distance between them has to be minute. If the two were to touch, not only would the result be stone grit mixed into the flour, but the valuable grinding surfaces would be damaged. In the Graeco‐Roman world, (p. 351 ) millstones were conical, the inclined grinding surfaces forcing the grind to travel across them under the pull of gravity. Developed for querns, they were also used at Hagendorn or the third‐century mill complex on the Gianicolo, indicating either low‐powered, slow‐moving millstones or a failure to grasp that with high‐powered, fast‐moving millstones the centrifugal momentum is sufficient to pull the grind across the grinding surfaces. Millstones in ancien régime Europe were exactly level and inscribed with an intricate pattern of straight or curving sharp‐edged lines. Wide‐profile lines brought in air to cool the grist and prevent it from being roasted during the milling process. Other lines broke up the grist and directed it along as long a route towards the outer rim as possible (or a less long route if groats rather than flour were desired). The lines needed to be regularly resharpened, and the surfaces between them roughened since the milling process polished them—unless one invested in high‐quality millstones of hard, small‐pored material, such as the burrstone of La Ferté‐sous‐Jouarre (on the Marne river east of Paris) or the lava stone from the Eifel hills near Mayen. As even prolonged wear would never render their surfaces smooth, millstones from those two places were so sought after that they could be exported widely, including to England—where in the pre‐Reformation period millstones from the Mayen area were known as ‘Cullen’ stones because they were marketed (via the Rhine) by merchants based in Cologne. ¹⁸

To keep the runner stone at just the right distance from the bedstone, a tentering mechanism was needed to permit it to be lifted and lowered with millimetre precision. It was in a raised position while the mill was being powered up; only when it reached the required rotational speed was it carefully lowered into the desired position, almost touching the grindstone but not quite. Keeping it in this position was somewhat tricky too, since its rotational momentum pushes it upwards. Since grist fed into the mill ‘lubricated’ the grinding surfaces as it was being ground up, it was imperative that the runner stone not turn in the lowered position when there was no grist to insulate it from the bedstone. In other words, the grist must not be allowed to run out while the runner stone was turning. This was ensured by the hopper suspended above the runner stone and feeding the grist into the hole at its centre through a 'shoe’ struck by the ‘clapper’, a damsel connected to the rotating machinery and first recorded in an illustration of about 1480. ¹⁹ Agitating the shoe and hopper was necessary to keep the grist flowing. For added safety, a bell would be mounted inside the hopper: it remained silent as long as it was covered with grist but, owing to the rhythmic shaking of the hopper, went off when it became uncovered. ²⁰

Perhaps because there were as yet few watermills in the first place, and perhaps also because the absence of overshot mills meant that less energy was available in total, mills in the Roman period appear to have been employed almost exclusively for milling grain. But with mills becoming a standard appliance, and especially with overshot mills in operation, a surplus of energy meant that more and more mills were employed for other purposes as well. Concerning the Roman period, a passage in Ausonius’ poem on the Mosel river of 371 is generally considered the only testimony of an ‘industrial’ application, beyond milling, of water power. In enumerating the tributaries of the Mosel, (p. 352 ) Ausonius says of one of them, the ‘marble‐famed Ruwer [marmore clarus Erubris]’, that

Evidently, the mills on the banks of the Ruwer were used both for milling grain and for cutting marble, and quite a lot of it too, it seems. Ausonius describes the rotation of the millstones as praeceps, a word that by its etymology literally designates a ‘headlong’ rush. This suggests that the mills in question were quite advanced. Millstones are not actually very loud, nor does their low rumble carry well, and wooden gear wheels, lubricated with tallow or beeswax, are practically noiseless. So the ‘ceaseless din’ must have been due to the saws eating into the marble. The same poem speaks of the grand country houses (villae) lined up along the Mosel, which, Ausonius says, had coffered ceilings of marble as well as floors of ‘Phrygian tiles [Phrygiae crustae]’, the raw material for the latter—popular in the Roman world—being quarried in Asia Minor (nor is marble found in the Ruwer area). The Ruwer joins the Mosel just downstream from Trier, at that time residence of the Roman emperor, Valentinian I, to whose entourage Ausonius belonged. The frequent presence of the ruler around the time when the poem was written may have caused additional demand both for flour and for upmarket building material. ²¹

In the second half of the first millennium, watermills clearly spread swiftly. The abbey of Saint Wandrille‐de‐Fontenelle in Normandy owned 63 in about 787, the abbey of Saint Germain‐des‐Prés in Paris had 85 around the year 820, and Prüm abbey in the Eifel hills had 50 in 893. ²² Those mills may still have been of a somewhat primitive type, like the Hagendorn one, or even horizontal mills. But the spread was important to create the critical mass that enabled the technology to progress: the more mill owners there were, the greater the chance that some improvement hit on by any one of them would be copied by his neighbours, and spread in its turn. The Domesday Book of 1086 lists some 6,000 watermills in the English kingdom. It assigns each a money value, which varies enormously—perhaps reflecting, among other things, different levels of technological achievement. ²³ Whereas until the eighth century more mills appear to have been given to monasteries than built by them, ²⁴ later on the role of the great monasteries in promoting watermills seems to have been considerable. From the twelfth century onwards, it was the Cistercians in particular whose rapid expansion throughout Europe meant a boost for water power technology. The Cistercians had a strongly ‘industrial’, engineering‐oriented streak, and liked to found their monasteries in remote areas, spreading state‐of‐the‐art mills to parts of the continent where they were perhaps not yet known. An unknown author presumably writing in the second half of twelfth century has left a somewhat (p. 353 ) humorous Latin description of the great abbey of Clairvaux, a model foundation and showpiece of the Cistercians and which established more daughter houses (some 350) all over Europe than any other of their monasteries, including Cîteaux itself. In a stunning display of verbal showmanship, the author devotes much space to the various production processes mechanized with the help of water power, illustrating the long way travelled since Antípatros. Whereas the Greek poet had only been aware of water power being used to mill grain, at Clairvaux, in the twelfth century, that was but one task performed by water power (though it still tops the list). Indeed the twelfth‐century text spells out a dependence on water power that to Antípatros was as yet unknown:

That part of the river which the wall, acting as porter, admits rushes first to the mill. Careful [i.e., anxious] and troubled about many things, ²⁵ now he crushes the grain under the weight of the millstones, and now separates the flour from the bran by means of a fine sieve [which, like the hopper, would be shaken back and forth by a damsel connected to the mill mechanism]. But already he is filling the cauldron in the next building, giving himself over to the fire for boiling so as to furnish a drink to the friars …[The allusion, elaborated on in the omitted part of the passage, is to the brewing of beer.] But he has not yet discharged his duties. For now the fullers, right besides the mill, are calling him over, demanding that just as in the mill he was careful [anxious] to provide nourishment to the friars, so in their workshop he by rights should see to it that they have garments, too. Nor does he object, or refuse any task that he is asked to perform: alternately, he lifts and lowers the heavy pestles, or, if you prefer, hammers—perhaps it is best to call them wooden feet, for that seems most appropriate to the fullers’ business, which is jumping [in non‐mechanized fulling, the fullers stomped or jumped on the fabric]. Thus, he acquits the fullers of their heavy labour, and, if a joke be permitted on so serious a topic, indeed acquits them of the penalty for their sins [expelling Adam and Eve from the garden of Eden, god, in the book of Genesis, puts this curse on them that as a punishment for disobeying him they and their descendants from now on have to toil for their livelihood]. Dear god, how much solace you offer to your poor children, lest they sink into excessive melancholy! How much relief from their punishment you grant those who are repentant, lest they risk being quashed under the weight of their labour! For how many horses broke their backs, how many human arms spent their strength in that drudgery which the river, without our doing anything, graciously takes from us, even as without it there would be no clothes for us, nor food? But the river shares himself with us, and expects no other profit of his labour which he taketh under the sun ²⁶ than that, having diligently accomplished everything, he is free to flow away. So he sends a great many nimble wheels spinning rapidly, and emerges foaming to the point of giving the impression of being ground and softened himself. Now he is received by the tanning house, where he displays a great deal of industrious activity in the making of what is necessary to produce [leather] shoes for the friars. Then, by and by dividing himself into many branches and busily flowing hither and thither, he has a look at all the different workshops, making a point of inquiring how he might be of service: whatever needs cooking, sieving, turning, grating, watering, laundering, milling, softening, his loyalty is unfailing. At last, unwilling to earn anything but total gratitude or to admit that his accomplishments fall short in any way, he leaves everything clean. ²⁷

(p. 354 ) The text illustrates that mills released a great deal of labour—which could then be employed elsewhere; at the same time productivity rose. The Greek and Roman world was not yet very good at exploiting water power—but without it there was no energy to power mechanized production processes. Contrary to a widespread perception, ²⁸ Graeco‐Roman civilization was not indifferent to technological innovation because it had slaves. Rather, with no other energy source available to power production it could not but rely heavily on human labour—which, unsurprisingly, fell first and foremost on the slaves as the least privileged part of the population, even though many craftsmen and farmers who were free men of course worked hard, too. Put simply, it was not that slaves rendered water power unnecessary, but that the lack of water power rendered their exertions the more necessary.

In fact, Roman civilization, too, went to great lengths to save muscle power—quite literally so. Prokópios, in the sixth century, was much impressed by the aqueducts serving the ancient capital of the empire: ‘[T]he aqueducts of Rome are fourteen in number, and were made of baked brick by the men of old, being of such breadth and height that it is possible for a man on horseback to ride in them.’ ²⁹ Figures on how long those aqueducts were are hard to come by and differ alarmingly—perhaps because the course of the aqueducts is not always quite certain, especially where they ran underground (as they normally did except when they crossed valleys or approached their end point), or because rather than taking account of every bend authors adopt a more pragmatic approach based on straight lines; generally the question seems under‐researched. Roughly speaking, the shorter kind were around 20 km (12 miles) long. The longest seems to have been the Aqua Marcia at a little over 90 km (56 miles), with the rest in between. Clearly, plentiful water at sufficient elevation could be found at the shorter distance; curiously, and contrary to what one would expect, there seems to be no correlation between the age of the aqueducts (the Marcia, completed in 140 bce, is one of the older) and the remoteness of their source.

Nor were long aqueducts the privilege of the capital. The aqueduct supplying Roman Cologne, on the very edge of the empire, is stated to have been 95 km (60 miles) long. ³⁰ Here, there may have been less choice—unlike Rome, Cologne is situated in a plain. A previous aqueduct carrying water from low hills closer to the city seems to have been lengthened in the second century to the more remote Eifel hills, presumably because the old source was deemed insufficient. The problem was to find an abundant source—in fact, the aqueduct draws on several—at a high enough elevation for the water to flow all the way to the city by itself. It was necessary to maintain a constant slight incline (at least overall—siphons were of course possible, e.g. to cross valleys): in fact even the new catchment area is not that remote as the crow flies, but the aqueduct made a considerable detour to avoid higher ground in between and cross the watershed between the Maas and the Rhine without having to dig through it at any great depth.

All this was done to obviate the need for pumps. We are accustomed to regarding the great Roman aqueducts with awe. In a way this is justified—they are great works of engineering. But at the same time they bear conspicuous testimony to the fact that (p. 355 ) Roman civilization lacked the knowledge to harness energy other than muscle power. Aqueducts were not built to obtain water as such. Both Rome and Cologne are after all situated on a great river, whose vicinity meant plenty of clean groundwater at shallow depth. Aqueducts were built to obtain running water. If no more aqueducts were built in Latin christendom, it was not because technology had regressed. If, as in the Roman period, pumps could be driven only by muscle power, then building 95 km of aqueduct might appear attractive. In Latin christendom, pumps would simply be powered by a water‐ or windmill, available at a fraction of the cost of an aqueduct. Water or wind power was frequently employed in Latin christendom and post‐Reformation Europe for pumping water—for example for irrigation, or for draining (of mine shafts, or of low‐lying land reclaimed from the sea). Where in the Roman period an aqueduct would have been the only solution, Louis XIV of France had the once‐famous (but now destroyed) machine de Marly, built in the 1680s: fourteen undershot waterwheels with a diameter of 8 m on the Seine river at Marly, driving 235 pumps that propelled water over a total difference in elevation of 165 m—their sole purpose being to feed the spectacular jets in the park of Versailles. ³¹

The clichéd depiction of the Graeco‐Roman world as lacking the interest in technology and innovation characteristic of the ‘early modern’ period may be due to a misperception of pre‐industrial technology as primitive and there for the taking. In fact, the technological evolution under review here probably followed its natural, exponential trajectory independently of any cultural paradigm shift: unhindered, that is, by any indifference or even hostility to technological innovation in the pre‐ or early christian period. The beginnings of watermilling were necessarily slow. Precisely because it was unaccustomed to watermills, the Graeco‐Roman world did not depend on them and thus necessarily failed to generate much pressure for perfecting the technology. A civilization in which the only machine exploiting non‐muscle power to be in widespread use was the sailing ship could have no notion of the impact that the kind of technological innovation represented by watermills might have. As long as watermills were few and far between, progress was haphazard, with technological dead ends (such as, it would appear, the Bagrada mills) inevitable.

Yet in the second half of the first millennium a tipping point was at last reached when the number of mills became great enough to spur rapid technological progress through emulation, comparison, and experiment. Now, the more people saw a watermill operating successfully, the more people wanted one for themselves. From this point onwards the evolution presumably became self‐reinforcing and ever faster. After the turn of the millennium, a veritable take‐off is evident, of which Clairvaux abbey is one manifestation. Waterwheels powered more and more applications: fulling cloth, tanning leather, boring holes, turning wood, cutting and sawing, polishing, whetting, grinding (e.g. for making gunpowder), and stamping (e.g. to soften and pulp rags for making paper in paper mills). Mills not only produced flour and bran, but also oil. Not least important, water power was ever more evident in metallurgy. Waterwheels were used to drain and ventilate mine shafts. Trip hammers crushed the ore as well as serving in forges; rolling mills mass‐produced sheet metal, wire drawing mills, invented in the early fifteenth century, mass‐produced wire. And waterwheels made possible the construction of blast furnaces.

(p. 356 ) Since the beginning of the iron age, iron ore had been smelted in small, charcoal‐fired ‘bloomery’ furnaces, oven‐like contraptions which were not hot enough actually to liquefy the metal. Rather, such furnaces initiate a chemical reaction that separates the metal by making the oxygen in the iron oxide of the ore react with the carbon monoxide building up in the oven to form carbon dioxide. The metal precipitates as blobs of ‘bloom iron’ of a dough‐like consistency, which has to be processed immediately to remove slag particles trapped in the bloom and give the metal a less compact form (such as rods or sheets) to facilitate its reheating to a temperature sufficient for forging it. By contrast, water‐powered bellows enabled larger furnaces consisting essentially of a huge chimney to reach much higher temperatures and produce molten iron. Such blast furnaces appear in Sweden, a country rich in iron ore, around 1350; by about 1500 they could be found in much of the rest of Europe as well. Bloomery furnaces took many hours to smelt a rather small amount of ore. Then they had to be restocked with fuel and ore and fired up again. By contrast, blast furnaces could be restocked while in operation and, as long as this was done, produced iron continuously and in much larger quantity.

At the beginning of the second millennium, iron was still scarce and precious in Europe: the smallest scraps were continually re‐forged, swords were precious heirlooms, iron ship's anchors were so expensive that, in twelfth‐century Venice, several merchants would join together to rent them. Blast furnaces changed that. It has been estimated that the total annual production of iron in Europe rose from 25,000 tons in 1400 to 40,000 in 1500 to 160,000 in 1760 (it would reach 20 million tons in 1900 and 100 million in 1960—a typical exponential curve). ³² Needless to say, water‐powered bellows also proved advantageous in smelting and forging other metal. Metal (ferrous or not) thus grew more plentiful and affordable, resulting in productivity gains in agriculture and many crafts. Metal had endless applications (e.g. horse shoes, iron ploughshares, other iron tools and gear, weapons, and armour). Blast furnaces also enabled iron canon balls to be mass‐produced, rapidly pushing stone shot off the market and helping the spread of heavy artillery.

The dynamism inherent in the exploitation of water power is also evident from the fact that it became an increasingly scarce resource. Good sites for new watermills were clearly increasingly hard to come by. By the late pre‐Reformation period, watermills dotted the banks even of the least suitable river. The river Niers (Neers in Dutch) is a good example. A typical river of the plain, the difference in altitude from its source near Mönchengladbach (a little to the west of Düsseldorf) to its confluent with the river Maas at Gennep just south of Nijmegen is no more than 67 m. Nevertheless, it powered a considerable number of watermills. Weirs were built to gain some head for them—but, unfortunately, reduced the head for any nearby mills upstream. The only solution was a code of conduct binding on mill owners along the entire course of the river. It was negotiated in 1487 between the princes through whose territory the Niers flowed—the duke of Guelders, the duke of Cleves, the duke of Jülich, and the archbishop of Cologne—and among other things laid down a maximum head for weirs of one and a half feet (Cologne measure, fractionally shorter than English feet), corresponding to 43.5 cm. This is rather little even for undershot wheels, and forced mill owners to build mill ponds: the ponds were left to fill up overnight or whenever the adjacent mill was not in operation, and emptied to power the mill. The rather draconian nature of this (p. 357 ) requirement, the fact that four generally quarrelsome princes managed to agree on it, and the fact that the 1487 mill code for the Niers was regularly renewed (the last time in 1841) indicate the strong pressure to have as many mills as possible on the banks of even a small, sluggish river like this.

Things were more comfortable a little further south on the river Erft, which between its source in the Eifel hills and its confluent with the Rhine just south of Düsseldorf drops 479 m. Whereas the incline of the Niers is only 0.52 per mille along its entire course, the incline of the Erft is 1.53 per mille even if its relatively steep upper course through the Eifel hills is not taken into account. Both rivers were canalized in the nineteenth century, shortening the course of the Niers by about 10 km and the course of the Erft by no less than about 30 km. Until then, they were practically equal in length at 129 km for the Niers and 131 km for the Erft (a little over 80 miles). But whereas, in the late eighteenth century, there were thirty‐nine watermills on the Niers, the total for the Erft was seventy‐five, so that the average distance between them was only 1.75 km (just over a mile). Whereas a mill code for the Niers proved indispensable already by the late pre‐Reformation period, it was only in 1772 that the two major princes on the Erft, the archbishop of Cologne and the duke of Jülich, named a joint commission to investigate the situation of the mills on that river. The commission found that on the Erft, too, weirs were a problem, and recommended that they should be allowed a maximum head of 3 or 3.5 feet, as opposed to the 5 feet or more frequently encountered. There seems to have been no follow‐up: evidently the problem was insufficiently urgent. ³³

The figures do indicate that watermills were spaced as close as possible, and no doubt often rather too close; many must have been of marginal merit. A map of mills mentioned in the 1086 Domesday Book shows much of England to have been densely covered in mills already at that date, with 30 mills along one 16‐km (10‐mile) stretch of the river Wylye in Wiltshire—but very few mills in northern England and Cornwall, though perhaps the 1086 record is incomplete for those parts of the country. ³⁴ Of course the saturation point where no more good sites could be found was not reached everywhere at the same time, depending on topography, degree of economic development, and the legal and political situation (a powerful mill operator could defend a monopoly for his mill—a 'suit of mill’ or ‘thirl’—in a relatively large district). Yet it seems clear that as Latin christendom became more and more dependent on watermills, the growing scarcity of good river or stream sites intensified the pressure for technological improvements.

This phenomenon presumably underlies the appearance of tide mills, at a surprisingly early date. They apparently existed in Ireland already in the seventh century. ³⁵ The Domesday Book lists one at Dover; at least 168 such mills dating from before 1500 have been identified in England and Wales, no doubt less than the total. ³⁶ This is the more remarkable an indication of the hunger for energy generated by the widespread adoption of watermills as tide mills are particularly difficult to build. The sites are usually tricky and exposed to destruction by storms, the flow reverses its direction twice daily (usually, therefore, there would be one mill wheel for the ‘inbound’ current and another for the ‘outbound’ current), and the constant variation of the water level means that the mills (p. 358 ) are operational only twice during daylight hours for a rather short period. But to offset this was the fact that the tidal flow on the Atlantic and North Sea coasts is very powerful.

Easier to build than tide mills were windmills—which, in England, not infrequently came to replace tide mills when those were destroyed in a storm, being a cheaper and safer alternative. ³⁷ They are recorded on both sides of the English Channel from about 1180 and presumably originated in that general region. Their power source, the wind, is far less reliable, and most of the time less strong, than either a good river or the tidal flow. But their advantage was that they could be built almost anywhere and that there was no limitation on the number of windmills on given sites. Unlike waterwheels, which competed with each other either directly (in the case of overshot wheels) or because their weirs took away head from the mills upstream, windmills hardly interfere with each other. Windmills were unknown in the pre‐christian Mediterranean world, and the kind first found in northern Europe (later southern Europe, too), whose vanes turn a horizontal or slightly inclined axle, appear to owe nothing to the type with vanes turning a vertical axle that emerged in the Middle East in the late first millennium. The windmills of Europe, it is worth underlining, were invented only because the spread of watermills had created a hunger for cheap non‐muscle kinetic energy, and because watermills (abstracting from the relatively few tide mills) could only be built along river banks, where the good sites were filling up or were inaccessible to those not living near the river. Windmills in Europe were a corollary of, and complemented, the watermills, to which they remained secondary; in particular they were ill‐suited for heavy machinery and for quasi‐industrial production processes that required a constant energy input. In that sense they are, above all, another testimony to the demand for energy generated by watermills.

The same pressure that led to the development of windmills from the twelfth century onwards may have been responsible for the development of overshot watermills as well, and perhaps at roughly the same time. If my hypothesis that contrary to the general assumption the overshot wheel was unknown in the ancient Mediterranean world is correct, then the question arises at what point in time it did appear. The earliest indisputable evidence for overshot mills that I have found is the depiction of one in a miniature dated to c.1225. ³⁸

The tendency towards growing mechanization manifest in Latin christendom has another aspect that should not be underestimated: it accustomed people to technological progress and made them look for it, including beyond the applications of water power that had started this tendency in the first place. The wiredrawing mill was invented, in the fifteenth century, at the instigation of the Nürnberg city council, which exempted investors in the project from the dues collected from mill operators. Beyond water power, the spinning wheel and the horizontal loom (both arriving from Asia in the thirteenth century) greatly increased productivity in textile production. Ultimately, inventions like printing and indeed the steam engine were the result of a mentality for which the exploitation of water power had first created the conditions—to say nothing of the fact that the enormous output of the printing presses of Europe, commonly estimated at fifteen to twenty million books in the four decades before 1500 alone, would likely not have been possible without the easy availability of inexpensive paper owed to water‐powered paper mills.

(p. 359 ) One intuitively understands that the Greeks and Romans could not have invented the steam engine, or if they had would not have known what to do with it: they lived in a society that being non‐mechanized (and low‐energy) had no need for (more) mechanization (or more energy) and was incurious about it. By contrast, and thanks, ultimately, to the simple watermills described by Antípatros or excavated at Hagendorn, ancien régime Europe would have been unthinkable without its machines. They literally were its daily bread: the flour for the bread that everybody ate every day came mostly from water‐ or windmills. Meanwhile, the labour for which Homer has the miller girl curse the suitors of Ithákê and which the twelfth‐century author of the description of Clairvaux abbey thanks god for delivering him from could be invested elsewhere and, often, in production processes that were likewise mechanized. Unlike the pre‐christian Mediterranean world, late ancien régime Europe was a civilization where the steam engine was not only more likely to be invented but also could not but catch on. Significantly, in the nineteenth century, watermills were sometimes converted by substituting a steam engine for the waterwheel to power the mill mechanism: they were not that far apart. It is not for nothing that in the English language industrial plant, even if no longer water‐powered, still goes by the name of ‘mill’, as in 'steel mill’ or ‘mill town’. ³⁹

4.1.2.1 The countryside ⁴⁰

Technology was not the only field in which the society of Latin christendom manifested a dynamism not found in the pre‐christian Mediterranean world. Even though subsistence remained the main economic aim of almost every household and community, the role of commerce gradually increased. Markets were not yet of fundamental importance for society. But they created the preconditions for the growing monetization of economic life after the turn of the millennium. That in turn caused social structures to change in a dynamic, cumulative, and self‐reinforcing development, of a kind unknown to the ancient Mediterranean world.

In the countryside, the growing role of monetized exchange and the growing demand for refined products and services that local resources could not provide but which were available for sale increased the demand for money. To give but one example, the typical lordly residence of eleventh‐century northern Europe, the ‘motte’—a moated artificial mound with a wooden tower and palisade on top, sometimes with a ‘bailey’, a fortified enclosure at its foot—could be created entirely from local materials and by drawing on labour services that the local population owed the lord. The much more refined stone castles that became ubiquitous from the twelfth century onwards could not. Peasants were no stonemasons (assuming that suitable stone could even be quarried locally), to say nothing of more complicated tasks required for constructing such edifices. ⁴¹

(p. 360 ) As a result there was a tendency for dues in kind to be converted into monetary dues. But monetary dues were subject to devaluation by inflation, which in the long term was often dramatic. To maintain the purchasing power of the revenue from such dues, they had to be raised and new dues introduced, neither of which was easy. One response to this difficulty was a trend to convert long‐term into short‐term leases so they could be put up more often and more easily. On the whole and in the long run, it was the peasants who benefited from inflation, since the market price of their produce would often rise faster than their lord could put up the dues that they owed. On the other hand, transforming labour services into monetary dues benefited the lord, too. Peasants rendering labour services had no incentive to work hard, and had to be fed by the lord for the duration, the cost of which might be considerable. Moreover, labour services required of the peasants were hardly ever very heavy, and were in fact often foregone.

As peasants acquired monetary wealth, they were differentiated by it. It was mostly the richer peasants that supplied the markets, whereas the poorer peasants were obliged to earn money by selling their labour. It was they who often left their village to join the swelling population of the towns or of the new settlements that in much of continental Europe sprang up in newly cleared woodlands. Contrary to a widespread perception, the mobility of the rural population in Latin christendom was high. Leaving the village was not especially difficult. Otherwise the population of the towns would not have grown as fast as it did (or at all, for that matter, since the mortality rate in the towns was persistently higher than the birth rate), to name but the most obvious effect of this mobility. ⁴²

As monetization progressed, many minor lords faced impoverishment, a phenomenon that by the end of the pre‐Reformation period had become widespread. ⁴³ Inflation reduced the real value of income from monetary dues, while at the same time the cost of the lifestyle that lords would consider appropriate kept rising—a lifestyle that owed much to the example of the richer burgesses of the towns, who, as we shall see in a moment, were at the centre of the monetization process. The great plague of the mid‐fourteenth century reduced the population of Europe considerably and lowered agricultural prices owing to lack of demand; they remained depressed throughout the fifteenth century. As the rural population decreased, too, and the towns remained a magnet for it (not least because salaries increased markedly and quite durably as a result of the plague), it was often difficult for the lords to find enough peasants to work the land. Indeed, monetary dues owed by those peasants might have to be lowered at a time when the financial situation of the lords required the opposite. Many minor lords therefore joined the service of greater lords for pay. Or they became ‘robber barons’—a phenomenon precisely of the fifteenth and early sixteenth centuries (though the term is not).

On the whole, the rural economy seems to have expanded in the period between the turn of the millennium and the early fourteenth century (stagnation set in already a generation or two before the advent of the great plague). Not only was the area under cultivation extended continually, especially with the great clearances of the twelfth and thirteenth centuries, but productivity increased, too: the fact that more grain was harvested was not only due to the larger area under cultivation but also to greater yields. (p. 361 ) This phenomenon is hard to pin down statistically, yet the general tendency seems clear enough. ⁴⁴

4.1.2.2 Long distance trade and the towns

Progress in agriculture benefited the towns, which could not have multiplied and grown as after the turn of the millennium they rapidly did had there not been the capacity to feed them. The towns played a key role in the process of monetization. They were very different from the towns of the pre‐christian Mediterranean world. Those towns had been dominated by landowners, with traders and craftsmen playing only a minor role. In the towns of Latin christendom, it was the opposite.

No more than the pre‐christian Mediterranean world was Latin christendom dependent on trade for its subsistence. Even in the towns each household tried as much as possible to remain self‐sufficient. Markets served to get rid of any surplus. But they did not control production, or furnish the bulk of what a household required. ⁴⁵ Many townspeople, even the less wealthy, bought agricultural rents, which played a major role in covering their basic everyday needs. Under this system, peasants pressed for cash borrowed money from townspeople in return for regular deliveries of agricultural produce—for example food, fuel, and wool. At first, rents could not be redeemed and were unlimited in time; later they could be bought back and tended to be limited to the lifetime of the creditor and his immediate heirs. (The lack of any obligation to pay back the principal meant that this kind of arrangement fell outside the prohibition of loans by the church.) ⁴⁶ Increasingly, too, townspeople would buy agricultural land and lease it out: again the payment might often be in kind. In seventeenth‐century Beauvais, a normal household would receive some or even most of its food, fuel, wool, yarn, or cloth, and thatch for its roof, from lease revenue in kind. ⁴⁷

Although they might use markets for buying and selling, few people depended on them, least of all on markets far away. The only merchandise playing a significant role in long‐distance trade that was not a raw material or comestible was cloth. But no one depended on imported cloth: it could even be woven at home. Other goods that were the object of significant long‐distance trade—such as timber, wax, metal, leather, fur, wine, salt, salted fish, and grain—could be had in many places and, often, locally, too, so that there was no dependence on given centres of production. Finished products for everyday use were rarely imported but made locally. ⁴⁸ To be sure, some great cities, or areas that were highly urbanized, might rely on long‐distance trade for part even of their basic needs, especially grain. Thus, in late Latin christendom much grain from the southern Baltic region was exported to Flanders, ⁴⁹ and the populous cities of Italy, such as Venice, Florence, and Genoa, also imported grain from far away. ⁵⁰ The Hansa (p. 362 ) used trade embargoes as a means to exert political pressure, which might cause food shortages—especially in Norway and Flanders. ⁵¹

But even here it is hard to speak of real dependence. Flanders in the early fourteenth century had an estimated 5.5 million inhabitants on the territory of what is now Belgium. That is more than half of the present population, at a time when Europe as a whole had perhaps a fifth of its present population. Twenty‐seven per cent of the inhabitants of Flanders are thought to have lived in towns: ⁵² they alone can have suffered from shortages of imported grain. Norway is a still more striking case in point. It is routinely stated that in late Latin christendom ‘Norway’, a main supplier of dried cod, relied on imported foodstuffs, especially grain. But ‘Norway’ can here mean only the population of the towns, of which there were four. Among them, Bergen with 8,000 inhabitants was the biggest by far, the other three not exceeding 2,000. ⁵³

The towns of Latin christendom kept attracting immigrants from the countryside. Many newcomers looked for employment in the crafts. The growing labour force in the crafts stimulated the competition between them and the creation of improved or new products. On the whole, owing to a low degree of mechanisation production remained labour‐intensive rather than capital‐intensive. Products were handled by the same worker throughout the production process.

Rather than division of labour there was a division of crafts bordering on the extreme. In Paris, around the year 1300, already more than 300 can be identified. In 14th and 15th‐century Frankfurt more than 200 have been counted. Thus, for example, joiners, turners, coopers, millwrights, and cartwrights split off the carpenters. The reason was failure to break up the production process; for individual craftsmen were rarely sufficiently skilled to be able to make very different products of equal quality. ⁵⁴

A worker lived in the workshop of a master craftsman more like a member of the household than as an employee in today's world. The relationship between the owner of the workshop and his labour force was thus primarily personal rather than mediated by money. The special position of the master craftsman was not so much the result of his possession of productive capital than of his recognition by the local guild—whose control over access to the market (the number of master craftsmen active at a given time was normally limited) served the purpose of securing the livelihood of all its members rather than fostering competitive entrepreneurial dynamism. With their survival ensured, individual craftsmen had no pressing need to expand their business, an ambition that in any case would often have been impeded by the narrowness of their markets: joiners, coopers, etc. were found everywhere and their products thus could not be exported. So although employment in the crafts went up as the towns grew more populous, this was not accompanied by much concentration of capital in the hands of few.

Trade was different. Whereas a craftsman primarily managed labour, a merchant primarily managed capital in the form of money and stock, which he could increase without any intrinsic limit. That is what led the society of Latin christendom onto a (p. 363 ) very different path of development from the ancient Mediterranean world, despite the fact that it was no more dependent on trade than pre‐christian civilization. The wealth of the (agrarian) elite of the towns of the old Mediterranean civilization had been based on a finite resource: land. By contrast, the wealth of the (mercantile) elite of the towns of Latin christendom was based on a non‐finite resource: capital invested in an expanding economy. If long‐distance trade became so much more important in Latin christendom than it had been in Graeco‐Roman civilization, it was not because it had become more necessary, but simply because the urban elite of Latin christendom consisted of traders not landowners.

Many burgesses came to own agricultural land when they grew wealthy, but unlike townspeople in the old Mediterranean world they were not wealthy because they owned land. Normally, when people first moved to the towns they did not own much at all. Instead, their wealth was generated by selling something, which they therefore had an interest in doing as much as possible. Nor, unlike the situation in the ancient world, was there anyone trying to prevent traders from rising to the top of the social hierarchy at least within the towns, since the great landowners mostly did not reside there and left the towns pretty much alone. Further, there was nothing to stop mercantile wealth from growing and the richer traders from leaving their less successful colleagues ever further behind—as a result of which by the end of the pre‐Reformation period we have the phenomenon, unknown to the ancient world, of super‐rich traders like the Medici or Fugger families, who operated across Europe and whose wealth was greater than that of most princes. Among traders, competition led not so much to differentiation of mercantile activity but to ever greater capital investment. This in turn stimulated the establishment of business relations regardless of distance: the more capital a merchant possessed already, the further afield he might have to look for opportunities to invest it to keep up with the competition of his peers. Wealth, high rank in the social hierarchy of the towns, and long‐distance trade tended to go together. Thus, in the towns of the Hansa (which put pressure on its members to adopt similar institutional arrangements) only the richest traders were ratsfähig, ‘eligible to the council’. ⁵⁵

As we saw, christendom was a single cultural community where no one sharing the faith could be considered an alien. Latin was spoken by all, including traders—as noted, it was the exclusive official language of the Hansa well into the second half of the fourteenth century. Nor was there much in the way of ‘protectionism’ from rulers: mostly, they wanted to attract traders not keep them out. It was therefore easy to extend business relations across christian Europe. The spread of the horizontal loom throughout christendom from the thirteenth century onwards provided a great stimulus to both the cloth industry and the cloth trade, on a continental scale and without precedent in western civilization. Much of the wool produced by English sheep was exported to Flanders. Much of the cloth produced there was exported to northern Italy for further processing (though much of the wool of northern Europe also went to Italy directly). The fine cloth of Italy was then in part re‐exported to northern Europe, while at the same time fine cloth was also produced in Flanders itself and exported; efforts were also made to build up a cloth industry in England. ⁵⁶ In the linen industry, the towns (p. 364 ) of southern Germany specialized in the production of fustian, a material made of both linen and cotton. The cotton was imported from the Near East, for example via Venice. The cloth itself was then sold all over Europe, and even re‐exported to Asia and North Africa. ⁵⁷ Cloth was thus responsible for a major part (in terms of value) of the sales in the long‐distance trade of the period, not least precisely because the geographical separation of different stages of the cloth‐making process meant that value was added more than once before the merchandise reached the consumer. This was not the case with products like wine, grain, timber, salt, and the like that were likewise hauled and sold across the continent, but without being processed again before they reached their final markets.

Growing production and growing long distance trade resulted in more and more money flowing back and forth across the continent. That in turn is mirrored in the evolution of supra‐local finance. Paying for goods imported from far away by sending bags of cash in the opposite direction was risky, and onerous. Coins differed across Latin christendom, so importing them from far away entailed the necessity to exchange them for local coins, at a cost. To be sure, some prestigious currencies existed that were accepted practically everywhere. They were used for payment in long‐distance trade; they were moreover necessary as a monetary standard by means of which local currencies could be compared, and as an accounting unit. Yet for traders in regular business contact with each other it made more sense to grant each other credit, only occasionally 'settling accounts’. Even then sums owed by one party to the other might be paid by means of a piece of parchment or paper constituting a title to someone else's debt. The chief instrument for cashless money transfer was the bill of exchange, which appeared in the thirteenth century; cheques and remittances, as well as clearing payments between banks, are found from about 1300. Payment in parchment or paper was not (or did not remain) limited to long‐distance trade. Pierre Goubert describes how even the local businessmen of seventeenth‐century Beauvais preferred a system of mutual credit to payment in cash, which occurred only rarely. ⁵⁸

A big advantage of this system was simply that it was safer than cash, since the documents circulating between business partners in lieu of cash were worthless to robbers (loss of such documents in long‐distance trade was guarded against by sending several numbered copies separately). But an even greater advantage was that this system of mutual credit put at the disposal of the traders more money than circulated in cash. The currencies of Latin christendom were based on gold or silver, precious because they were rare; hence, cash was rare, too. If cash had been the only means of payment, no more goods would have been offered for sale than cash existed to buy them with (if we abstract from barter trade); and that would have meant far fewer goods reaching the markets than could be produced. To the extent that trade was financed from credit, the monetary mass increased beyond the supply of cash based on precious metal, and this was of fundamental importance as an enabling factor for trade.

Traders could grant each other credit back and forth without paying much heed to the limited availability of coined precious metal. Since only a fraction of it was ever actually transformed into cash, the total credit volume in christendom could be far greater than (p. 365 ) the total amount of cash in circulation. At his death in 1693, Gabriel Motte, one of the richest traders of Beauvais, left a monetary fortune of 163,000 livres. Of that amount only 4,000 livres (less than 2.5 per cent) were in cash, the rest in paper ⁵⁹ —which gives some idea of the dimensions of the phenomenon. But an augmentation of the monetary mass caused more products to be offered for sale, and the economy to grow. By contrast, in the old Roman empire, where there was little in the way of supralocal credit, the limited availability of cash acted like a straight‐jacket on the economy. Debasing the coinage, the remedy resorted to by third‐century Roman rulers, also increased the monetary mass, but payment in bad currency did not spur production.

Well aware of the growing role of credit for trade, the towns of the Hansa from the late thirteenth century onwards kept municipal credit ledgers as a service to merchants, both enabling them to have their claims to outstanding debts registered officially and publicly, and facilitating clearing operations (similarly, in seventeenth‐century Beauvais debts and claims were authenticated by the civic authorities, or a notary). Indeed the credit volume went up so quickly that already in the fourteenth century attempts were made by the Hansa to reign in the expansion of credit‐financed trade, in particular by trying—with only moderate success—to exclude credit operations involving business partners outside the Hansa. In the fifteenth century, maritime insurance developed in the Mediterranean to protect merchants from the loss of cargoes. In the Hansa, this was frowned upon, and the old method of dividing up cargoes among many owners (e.g. four, eight, or twelve Parten—parts—belonging to as many Partner or partners) clung to. Splitting the risk in this fashion was also an efficient protection but unlike the new Italian method involved no financial operations and had no effect on the monetary mass. Even bankers were frowned upon by the Hansa: no banking firms developed within it, although the Medici and other Italian companies offering financial services had branches for example in Lübeck. It is tempting to see this kind of conservatism as a possible cause of the decline of the Hansa that became manifest in the sixteenth century. ⁶⁰

In the thirteenth and fourteenth centuries, Italian banking firms were heavily engaged in financing the English crown. Philip IV of France also had Italian bankers: around 1300, the brothers Albizzo and Musciatto Guidi, representatives of a Florentine trading company, were so conspicuous at the French court that they were popularly known as Biche et Mouche (‘hind and fly’). Italian bankers contributed to financing the papacy, for example by participating in the collection of ‘Saint Peter's penny’, a levy raised throughout christendom and destined for Rome: in the fifteenth century, the Scandinavian revenue from this source was channelled via the Lübeck branch of the Medici and other Florentine firms in Lübeck. ⁶¹ In the fifteenth and sixteenth centuries, trading and banking firms like, in Augsburg, the Fugger family (first mentioned in the fourteenth century as weavers and cloth merchants) and the Welser family loaned huge amounts of money to the Habsburg dynasty, while in Florence the Medici became rulers of their city and were at length accepted as princes by the royal houses of Europe (in Germany, the Fugger family also rose to princely rank). ⁶²

(p. 366 ) 4.1.2.3 Monetization and political change

Unlike pre‐christian Greek and Roman society, in Latin christendom town and country were clearly distinct in legal terms. In the old Mediterranean world, the land surrounding a town (and usually extending all the way to the borders with the land belonging to neighbouring municipalities) was as much part of that town as the central square. There was, thus, little inhabited land that was not legally part of a town. By contrast, in Latin christendom a town was clearly marked apart, both in terms of legal status and territorially, from the surrounding countryside. Here, most land did not belong to a town but was part of an entirely different legal sphere. A town had ‘freedoms’, privileges exempting it from that rural sphere, into which it had developed as a legal newcomer and within which it constituted an enclave.

In the old Mediterranean world, where trade was of limited importance, landownership determined wealth and status. Divided up among the existing citizens, the territory belonging to a town was difficult to extend for reasons discussed in Chapter 2 . As a result, the citizens had no incentive to accept new arrivals in their midst, or to give much encouragement to resident non‐citizens active as craftsmen or merchants: if such non‐citizens became too prosperous, they might have gained too much influence, or at least made uncomfortable demands, jeopardizing the monopoly on landownership of the citizens. In a fundamental sense both the economy and the politics of a town of the ancient Graeco‐Roman world were zero‐sum games. A town in Latin christendom—after the turn of the millennium—was different. Here, neither an increase of the total number of citizens through immigration nor the accumulation of wealth in the hands of individual town‐dwellers had a negative impact on the community. On the contrary, they strengthened it. Wealth was not based on a fixed resource (land), but on the production of goods and services, which in principle could be expanded endlessly. United in solidarity against the rural world of the lords beyond its walls, the citizens (burgesses) had an interest in being as numerous as possible, not in keeping migrants out or at least down.

Yet the greater lords also had an interest in seeing the burgesses prosper. Some early towns of the new type arose spontaneously as merchants and craftsmen congregated for example near a royal or episcopal residence. Many more towns were founded at the initiative of powerful nobles, who gave ‘their’ towns privileges exempting them at least partially from their rule. This was because the towns attracted money, goods, and skills that in purely rural markets would not have been available even to the lords. At least as importantly, the lords profited for example from toll revenue along the roads, rivers, and harbours used by the traders. As a result, nobody was opposed to the growth of individual towns or the growth of the number of burgesses within them or the growth of the total number of towns, since everybody profited from this expansion. In terms of game theory, this was a highly dynamic win‐win situation.

The whole phenomenon was unprecedented in the history of western civilization—and it had important consequences for the evolution of its political structures. Whoever was able to ‘tap’ the growing flows of money between the markets of christendom gained a decisive advantage in the competition for political power. The late pre‐Reformation period was a time of rapidly growing monetary dues—commuted service obligations, tolls and fees, legal fines, and taxes. They increasingly replaced (but also were simply added to) older labour dues and dues in kind. For levying such monetary dues, especially taxes, secular lords were better placed than the church. It was not clear what the church (p. 367 ) could now 'sell’ people that it had not offered before. Whereas secular lords provided protection and law courts of which it could always be argued that more was needed, the church could not offer more complete salvation than it had, or claim that the traditional dues payable to it, such as tithes, were no longer adequate. To be sure, it did introduce new 'services’—most famously, granting indulgences (that promised faster salvation by reducing the time a person had to spend in purgatory to atone for his or her sins) for a fee. But there were narrow limits to what it could do to profit from monetization.

In Latin christendom, money taxes had fallen into disuse after the end of the old (west) Roman empire, and re‐establishing them after the turn of the millennium was a slow and incremental process. Secular lords therefore turned to other, easier sources of money, such as fees, legal fines, tolls, mining, and minting. The right of coinage, that is, the right to produce the coins that, as a consequence of the growth of markets, were in ever greater demand, could be an important source of revenue. Coins were stamped with the official emblem of the emitter, who in this fashion certified their metal value and vouched for it with his reputation. As pope Innocent IV in the mid‐thirteenth century recognized, this was a service justifying the fact that such coins circulated at a higher nominal value than their metal value, with the emitter pocketing the difference. Even professional money changers could determine the exact precious metal content of given coins only by melting them down; the guarantee implicit in the emblem of the emitter normally made this unnecessary. Besides, coins with a high precious metal content were soft and not very resistant to wear; bits along the rim would even be cut off deliberately to gain precious metal, in spite of prohibitions. Coins used for payment would therefore be weighed rather than counted (the names of many currencies—such as ‘pound’, livre, lira, Mark, peso, peseta—are actually the names of units of weight). ⁶³

Crucially, the more money a ruler could extract from his (rarely her) subjects the better able was he to establish effective supralocal authority. This sounds banal, but it is not. For this to work two conditions had to be met that, in Latin christendom before the thirteenth century, were in fact met only in the most inchoate fashion: the economy had to be monetized, and the ruler had to be able to secure a share of the money circulating among his subjects for himself. Let us remind ourselves of two remarks by Marc Bloch cited in Chapter 3 and which characterize the situation of Latin christendom before the thirteenth century. Là où les transports sont difficiles, l'homme va vers la chose plus aisément qu'il ne fait la chose venir à lui—where transport is difficult it is people who travel rather than the things they want. And: seul le pouvoir sur place était un pouvoir efficace, only local power was effective power. ⁶⁴ This was because such power was based on a right on the part of the ruler to services and dues in kind rather than money. Services and dues in kind could be consumed only at the place of their production. To be sure, goods could be shipped to places far away, and those who owed services could be made to travel, but both were practical only to a limited degree. Transporting goods was laborious, and scarcely possible if they were perishable. People who owed services could not be kept away from their homes for very long. Moreover, the ‘revenue’ from dues in kind or services could not be accumulated, at least not very well: stockpiles were again not possible if goods were perishable, and might not be very useful anyway since not only (p. 368 ) did the goods have to be transported to central storage facilities, but then transported again to where they might be needed. Likewise, a standing army was impossible to build from people whose obligation to serve was limited to six weeks per year.

Hence, until the thirteenth century or indeed later, the ‘travelling kingship’ of Germany, the Scandinavian and Iberian realms, Scotland, or England (already in the Anglo‐Saxon period, and still more pronounced in the Norman and Angevin period, when the royal dominions comprised vast continental possessions in addition to England itself): the king and his court travelled from one estate of the royal demesne to the next to profit locally from the goods and services due to them. Similarly, the French monarchy was essentially limited to the region around Paris and Orléans because that was where the royal demesne was concentrated. Monetization brought massive change to this system. Replacing labour dues and dues in kind with monetary dues had evident advantages for everyone. Even the local substitution of money dues for other dues increased the circulation of money in the economy as a whole and enhanced the ability of great lords to siphon off some of this wealth for themselves. Money even in the form of coins (rather than parchment or paper) was relatively easy to transport, it was not perishable and could thus be accumulated, and it could easily be converted, at whichever time was convenient, into most other things its possessor might need or want. The ruler could now stay where he was (increasingly, a fixed central residence), and yet exercise control even over faraway places by means of paid agents of often relatively humble rank. They were much more dependent on him than vassals (who of course did not disappear overnight, indeed the ‘feudal’ system was not abolished anywhere in Europe before the end of the ancien régime).

The observation that, after the turn of the millennium, kingdoms became less fluid, indeed virtually indestructible if they possessed an established, recognized crown—no matter if that crown was actually weak—runs counter to established ‘Realist’‐style wisdom, which posits that 'states’, certainly in that period, survived because of their ability to defend themselves. ⁸⁴ But it is borne out if we look, for example, at the British ‘Celtic fringe’: lacking a crown, Ireland and Wales were easily satellized by the Plantagenet dynasty of England, whereas Scotland escaped this fate. Nor was the English crown, in the ‘Hundred Years War’, able to absorb the French crown.

4.2.3.1 Ireland and Wales I: the absence of unitary native kingship leads to absorption by the English crown

If Ireland, in the twelfth century, became nominally subject to the English crown without even having to be conquered, it was ultimately because there was no Irish (p. 379 ) kingship comparable to that of other European kingdoms. Like a microcosm of Latin christendom at large, Ireland was notionally organized as a social hierarchy encompassing the entire island and whose politically important figures, at every level, bore the title rí (rig[h] in more ancient spelling, plural ríthe). An Indo‐European cognate of Latin rex, the word is ordinarily if misleadingly translated as ‘king’. ‘Leader’ would perhaps be more appropriate (certainly, Latin rex is related to the verb rego ‘to lead’), or simply ‘lord’. ríthe were elected from among the family of their predecessor, usually while that predecessor was still alive. On the local level, there would be a rí túaithe, the leader of a túath (plural túatha). A cognate of Germanic theod ‘people’, this word designated a (perhaps somewhat nominal) community of descent, a clan; the number of túatha has been estimated at around 150 in Ireland as a whole. At the next higher level, a ruiri was presumably a rí to whom other local ríthe owed allegiance. The term rí ruirech would appear to mark someone ranking at the level above that of ruiri. At the top of the hierarchy there was the ard rí, a term habitually translated as ‘high king’. After the turn of the second millennium, this title designated a primacy of honour among the great ríthe attributed to one of them by his peers. It seems to be a relatively late development, even if the sources trace it back to a mythic past; nor was this position always filled. Generally, superordination in the Irish hierarchy did not entail any formal power of command. ⁸⁵

In 1155, pope Hadrian IV, himself an Englishman, granted the English king permission to land in Ireland (ut …insulam illam ingrediaris) in order to reform the morals of the Irish (pro corrigendis moribus) and on condition that the king would levy in Ireland an annual tribute to the Holy See. ⁸⁶ The text, known as the bull Laudabiliter, states that the English king had requested it, but it is not clear whether the initiative really came from the English crown or whether, for example, church circles were behind it. The reference to Irish morals may not have been entirely hypocritical, since, among other liberties with church doctrine and to the chagrin of the clergy, divorce for example was at that time common in Irish society. For his part, the English king, Henry II Plantagenet, was evidently in no hurry to invade Ireland since, following Laudabiliter, he did nothing of the sort, not even when, in 1167, the rí of Leinster, Diarmait Mac Murchadha (Dermot MacMurrough), called on him for support in a quarrel that opposed him to the then ard rí. Diarmait politely did homage to Henry, but rather than intervene himself the king merely authorized Diarmait to recruit Anglo‐Norman barons from the unruly Welsh marches for his cause.

However, those barons were so successful militarily that in 1171 the king at last did land in Ireland to receive the homage of the Irish magnates, including the ard rí, a title that was henceforth redundant. ⁸⁷ It seems that, just as much as Henry himself, the Irish feared that otherwise the immigrant barons would grow too powerful. Under feudal law, their allegiance to the English king made him the protector of the Irish and thus made it more difficult for the barons, out to win land for themselves, to attack them. The Irish may also have regarded their acknowledgement of the English king as a natural prolongation of their own hierarchy, since, after all, they considered themselves part of (p. 380 ) the larger christian community, too: ‘The Irish had always recognized the Roman emperor [i.e. the German ruler] as “king of the world.”’ ⁸⁸ So did Henry, who, in 1157, had done homage to his ‘most cherished friend [precordialis amicus]’ the emperor, at a time when, king of England as well as lord of much of France, he himself was probably the most powerful secular ruler in Latin christendom:

We put our kingdom …at your disposal, so that everything may be done in accordance with your wishes and in conformity with your imperial will. Let there be indivisible unity of peace and love between us and our peoples …albeit in such a way that you, being of higher rank, have final authority, while we shall not be found wanting in obedience. ⁸⁹

This had no practical effect whatever, and no doubt was not expected to. The Irish may likewise have felt that their homage to Henry would at best provide some protection from the barons, and at worst remain an empty gesture.

No compulsion against the Irish was exercised by the English king during his stay on the island. He had the support of the Irish episcopate, who may have hoped to gain, with his help, a stronger position for the clergy in Irish society, similar to that of the clergy in England. Pope Alexander III confirmed Henry as secular master of Ireland in letters to the king as well as to the secular and ecclesiastical magnates of Ireland, expressly describing the submission of the Irish as voluntary. ⁹⁰ Pope Hadrian in Laudabiliter had referred to Ireland as an island, as a country (terra), and as a nation (gens), but not as a kingdom (regnum). Accordingly, Henry took the title ‘lord of Ireland’ (dominus Hiberniae), not ‘king of Ireland’. It was only in 1542 that Henry VIII adopted that latter title.

Like Ireland, Wales, after the end of Roman rule in the early fifth century, had always been divided into plural principalities. In Welsh sources, their leaders are sometimes called brenin, which in present‐day Welsh is the equivalent of the English word ‘king’, as well as rhi (the etymological counterpart of rí in Irish). After the turn of the millennium, Welsh sources prefer tywysog, equated with Latin princeps ‘prince’, and arglwydd, equated with Latin dominus ‘lord’, whereas English sources written in Latin actually tend to employ rex. ⁹¹ As in Ireland, we have the phenomenon of voluntary submission of one of the Welsh magnates to the English crown as a result of a power struggle within Wales: in 1267, the prince of Gwynedd, Llywelyn ap Gruffudd, recognized the suzerainty of the English king, Henry III Plantagenet, in return for support against Llywelyn's two brothers, and in order for Llywelyn to gain the overlordship of the whole of Wales. Henry III in his turn recognized Llywelyn as princeps Wallie, ‘prince of Wales’, a title originally adopted in a treaty between Llywelyn and Simon de Montfort, the leader of the English rebels against Henry in the so‐called Barons’ War, and later approved by Henry. If homage done to the English crown by Welsh magnates was nothing new in itself, the adoption of a title implying rulership of one of those magnates over Wales in its entirety was. Subsequently, however, Llywelyn refused to pay homage to the new English king, Edward I, despite (p. 381 ) being called on to do so several times. Edward thereupon marched against Llywelyn and defeated him.

Llywelyn kept the title princeps Wallie, but his authority was once more largely limited to his native Gwynedd. Moreover, he was left in a very uncomfortable position, suspect both to the English and to the Welsh and with little leverage over either. When a (minor) Welsh uprising broke out, Llywelyn had no option but to join it, the more so as the rebellion was led by one of his brothers—if standing back might have been safer, it would also have meant sacrificing all credibility with the Welsh. Llywelyn was killed somewhat accidentally in a skirmish in 1282. It was only thereafter that the Welsh‐speaking territory was largely occupied by English troops, partly converted into shires placed under royal administration (but often entrusted to indigenous nobles), and secured with a number of strong castles.

4.2.3.2 Scotland I: a unitary, but non‐standard native kingship

We shall return to Ireland and Wales but must first turn to Scotland. Whereas, in the first millennium, Ireland and Wales were ethnically quite homogeneous, this was not true of northern Britain. The inhabitants of the north and west of what was to become Scotland, known as Scot(t)i, were Irish by language and culture. Much of the east was populated by the Picts. The area around Glasgow (Strathclyde) belonged to Britons speaking the language that in English is now called Welsh, and the south and the area around Edinburgh (Lothian) in the latter part of the first millennium were settled by Angles. This ethnic diversity may at least partly account for the fact that conflict between those groups of inhabitants seems to have been fiercer than conflict among the inhabitants of Ireland and Wales. At any rate, it resulted in the unification of Scotland by the rulers of the Irish realm of Dál Riata straddling Northern Ireland (Antrim) and western Scotland (Argyll) and inhabited by the Scot(t)i. With what appears to have been considerable brutality, Cináed Mac Alpín (Kenneth MacAlpin, reigns 841–58) of Dál Riata destroyed the Pictish principalities already weakened by Viking raids.

Later chroniclers regard him as the first incumbent of what was to become the Scottish throne, even though it was not before the early eleventh century that both the Britons of Strathclyde and the Angles of Lothian had finally been brought under ‘Scottish’ rule: all rival centres of power having been eliminated, the appellation Scot(t)i came to be applied to all the subjects of Cináed's successors. Based on conquest, this Scottish kingship was evidently stronger, in organizational terms, than those whom it subjugated, but it was weaker than Norman kingship, likewise based on conquest. After his occupation of England from 1066 onwards, William of Normandy advanced with his army and navy to the Firth of Tay, but omitted to conquer Scotland. Moreover, the Scottish king Malcolm III (reigns 1058–93) did homage to William. Thereafter, relations between the two kingdoms remained largely friendly, while culturally as well as politically Scotland was exposed to strong Anglo‐Norman influence.

Although, from today's perspective, the ‘feudal’ system seems to be characterized by decentralization of power, it was nevertheless more centralized, and focused on the king, than the Irish system, where the lower ríthe were not in any formal way linked, let alone answerable, to those of higher rank. Until the twelfth century, the Scottish system, with its Irish cultural heritage, seems to have been quite similar to that Irish system in that the leading nobles were not in any formal way obligated to the king. To reduce the (p. 382 ) influence of the native magnates, king David I (reigns 1124–53) enfeoffed Anglo‐Norman barons like the (de) Bruce (who took their name from the town of Brix in Normandy) and the (de) Balliol (who called themselves after the town of Bailleul‐en‐Vimeu in Picardy) with estates, court offices, and the new Scottish bishoprics and abbeys that he founded—that latter measure designed to reduce the influence of the native Celtic church, which like the native magnates had no formal ties with the king. However, at least some of the native nobles adapted, and likewise became royal vassals. We have the same phenomenon here as in contemporary Germany, with the allegiance of some subcommunity within the realm being replaced by allegiance to the king as the main ingredient of the legitimacy of the magnates. David also founded towns, with settlers in part recruited in the Low Countries and Germany, which became an important source of revenue for the Scottish monarchy.

David in this fashion did much to turn Scotland into a ‘normal’ kingdom similar to those found elsewhere in christendom at that time. Yet it remained archaic in one important respect. Like Irish ríthe, and due evidently to the Irish cultural legacy of the realm of Dál Riata, the Scottish king was not crowned or anointed at his inauguration. Until purged by David, this seems originally to have involved much ancient ‘pagan’ ritual, like the sacrifice of a horse. Typically Irish elements of the enkinging that were retained even by David included reciting the genealogy of the king at great length (it went back to Scota, daughter of Pharao) and in the Irish language—Latin, the ‘holy’ language of the church and of ‘modern’ administration, was probably not used in the ceremony. The new king was invested with a staff and perhaps a robe, and the ceremony somehow (the details are now lost) involved a sacred stone, a plain, roughly hewn slab on which the king may have stepped or sat. This sacred stone is often referred to as the Scottish ‘coronation stone’, a misnomer since the enkinging involved no coronation. Scone having been an important religious centre of the Picts, it was there that the ceremony was held—at least this was certainly the case after the foundation of Scone abbey in 1115. However, the event did not take place in the abbey church but next to it, in the open air—certainly this was so at the inauguration of Alexander III in 1249, which is the first such occasion about which we know even this much.

In a basic sense, this is the type of enkinging preferred over the more ‘modern’ coronation and unction by Henry I in 919. But Henry had the choice: the more ‘modern’ mode was offered to him, and his successors, beginning with his son Otto in 937, all availed themselves of it—to the point that after the turn of the millennium nothing else was even conceivable any longer. The Scots, for their part, by the thirteenth century would dearly have liked their king to be properly crowned and anointed. The catch was that this required a bishop, and he needed papal authorization. It was repeatedly petitioned for by the Scots, but Rome always turned them down, in response to English objections based on a claim that the Scottish rulers were vassals of the English crown. ⁹²

In 1286, the royal house unexpectedly became extinct. The regents arranged for the future marriage of the heiress to the Scottish throne, 3‐year‐old Margaret of Norway, with the English crown prince (himself only aged 2 at that time), while the English king, Edward I, agreed that Scotland and England would never be merged even if they came to share the same king (treaty of Birgham). But Margaret died during her crossing to (p. 383 ) Scotland. This left no fewer than thirteen claimants to the Scottish throne, all more or less distantly related to the defunct dynasty. The Scots sought the arbitration of Edward I, who agreed to examine the respective merits of the claimants but insisted that the Scots had to acknowledge his suzerainty over the Scottish kingdom; this was reluctantly agreed to. After all, the previous king Alexander III had also done homage to the English king, although the Scots liked to think that this was only for his possessions in England (I do not know what, in Alexander's case, those possessions were, but David I held three English counties from king Henry I, his brother‐in‐law). In what seems to have been a scrupulously legal procedure, Edward decided that of the thirteen candidates John Balliol had the best dynastic claim. John duly became king and did homage to Edward. If this homage had been treated as essentially symbolical like that of the Irish magnates to Henry II or that king's homage to the emperor, history would have taken a different course. But Edward, a powerful and energetic ruler, was not content with the token overlordship common on every level of the feudal hierarchy. He encouraged the Scots to appeal to him, the supreme liege lord, against their king, and imposed humiliating sanctions when John objected. On a strict interpretation of feudal law, Edward was within his rights, but politically his behaviour was a disaster for John.

In his desire to strengthen his kingship by making the most of his formal prerogatives, Edward was much like his contemporary king Philip IV of France, whom we encountered in Chapter 3 . Philip in fact made the same sort of demands of Edward in his capacity as Edward's liege for the continental possessions of the English king as Edward was making of John. When Edward failed to comply, Philip in 1294 stripped him of the duchy of Aquitaine. Edward refused to give it up, which meant war with Philip. As any liege faced with a feud would have done, the English king called on his vassals to serve him in this war with their men, including king John. For John it was the last straw: he could not ask the Scots to follow him into a war in a foreign land on behalf of the English king and retain any hope of acceptance of his own kingship by them. Instead, in a desperate bid to save his throne, he entered into an alliance with Philip against Edward. Edward thus had a perfectly legitimate cause for war on John as well as Philip. Well‐organized militarily as he was, he forced John into submission in a matter of weeks, formally stripping him of his kingship and putting him under arrest. The Scots were disorganized and disunited. Support for John was lukewarm, and some Scots indeed supported Edward even on the battlefield—among them the two Robert Bruce, father and son, whose claim to the Scottish throne was considered the best after that of John Balliol and who must have hoped that, John having failed, Edward would now give the Scottish throne to them.

But he did not. Edward may have felt that such a move could only lead to the same unsatisfactory situation as before, where he could not make any demands in his capacity as liege of the Scottish king without undermining the position of that king in Scotland itself. He could, then, either allow his suzerainty over Scotland to become a mere empty claim, or take the country over himself. Given his personality, and the situation (he had just, more or less, conquered the country), it is not surprising that he chose the latter. On the other hand, if the Bruce family expected that Edward would name another native Scottish king, it was because anyone could see that the attempt by Edward to impose his authority in Scotland directly was ambitious to the point of temerity. After all, he was a foreigner whose sole claim to the Scottish throne was a suzerainty that the Scots themselves had always tended to contest, and only half‐accepted as long as it remained (p. 384 ) largely meaningless in practice. Edward must of course have been aware of this problem, and this may explain why, having chosen the option that for him was the only acceptable one, he made no attempt to placate the Scots. Instead, he seems to have decided quite consciously that ruthlessness was the only viable strategy. This included dismantling Scotland as a kingdom.

As liege, he would have been entitled to keep the Scottish throne, which had escheated to him, for himself, simply adding the Scottish kingship to the English kingship that was his already but without merging the two (had the treaty of Birgham, voluntarily agreed to by the Scots, been implemented, this would have happened in the next generation in any case). But the Scottish kingship, defined, like any other European kingship, by a specific tradition and set of customary rights, was weaker than the English kingship, and so this solution would have demanded of Edward, the powerful ruler of England, not to employ the resources at his disposal in that country to impose a claim to the kingship of Scotland so controversial that those resources were actually badly needed. This would have been schizophrenic; small wonder that Edward, though by no means unscrupulous with regard to legal questions, in this situation chose resources over legitimacy and attempted to hold Scotland by force. Besides, I suspect that for Edward the idea of going to Scone and subjecting himself to a ritual whose christian credentials were at best dubious, and performed in a language he did not understand, was too abstruse to be entertained. What he did instead was take the sacred stone from Scone Abbey to Westminster Abbey, where he had it incorporated in a specially made throne. This so‐called Coronation Chair is still kept at Westminster Abbey and has been used for the coronation of subsequent English/British monarchs down to Elizabeth II in 1953 (in 1996, however, the septcentenary of its being brought to England, the stone was returned to Scotland).

The case of Scotland is similar to that of Wales in that there, too, following the failure of the attempt to exercise power indirectly in his capacity as liege, Edward assumed direct control and secured it militarily, this being the only alternative to becoming prince of Wales himself. The latter option was not really acceptable since as with Scottish kingship, though for a different reason, the position was not suitable for Edward. Whereas Scottish kingship was sanctioned by tradition but of inferior status, the title ‘prince of Wales’ had only been created for Llywelyn, and had never had real backing among the Welsh at large. The takeover of Wales by the English crown worked because however much the English might be disliked the Welsh had never been united. By the time they came under direct English rule, it was too late to create a common polity to serve as a framework for successful resistance. By contrast, the English failed in Scotland even though the full weight of English military might was brought to bear on Scotland in a way that Wales, let alone pre‐Reformation Ireland, never experienced. There were many contingent factors, even chance events, that contributed to the successful assertion of Scottish independence. But the crucial difference went far back in time and ultimately lay in the unification of Scotland by Cináed Mac Alpín and his successors, and the efforts, in particular of David I, to bring Scottish kingship in line with the prevailing political culture of the age. By the turn of the fourteenth century, it was too far advanced on the road to consolidation for Scotland to go the way of Ireland and Wales and become annexed to the English crown. Even such as it was, Scottish kingship provided too clear a focal point for what had, unlike Ireland and Wales, become a unified polity in its own right, to be swallowed up even by so powerful a kingdom as England was at that time.

(p. 385 ) 4.2.3.3 Scotland II: a unitary native kingship reasserted and standardized

This manifested itself early on in the immediacy, scale, and degree of organization of the Scottish rebellion. In pre‐Reformation Ireland, there was no large‐scale contestation of English suzerainty at all. In Wales, resistance was sporadic and mostly low‐key until the great revolt of Owain Glyn Dŵr in the early fifteenth century. Ten years passed between the reorganization of the country by Edward in 1284 and the brief Welsh revolt of 1294/5. By contrast, the Scots rose against Edward the moment he left the country again, and inflicted a resounding defeat on the English in the battle of Stirling Bridge (1297). To be sure, this was no doubt due partly to the English crown being taken by surprise at the quickness and magnitude of this reaction, and the initial Scottish triumph was followed by a long series of military disasters. Yet despite their defeat at Falkirk (1298) the Scots went on to form a regency council.

The Scottish resistance was not powered by nationalism in an ethno‐cultural sense. At its forefront were leading families of Anglo‐Norman origin who held estates and fiefs in England as well as Scotland. They were culturally indistinguishable from the political elite of the English kingdom, or indeed from the very barons who had also partly colonized Wales and Ireland. Thus, the Balliol family had vast estates in northern England and had founded a college at Oxford. The Bruce family likewise had important English estates; Robert Bruce junior, subsequently king of Scots and hero of the resistance, was married (in his second marriage) to the daughter of a powerful figure at the English court and holder of an Irish fief, Richard de Burgh, earl of Ulster. But the Scottish nobles had become attached to their own court and king and the prestige and preferment emanating therefrom. They were not going to have that taken away from them without a fight, especially as the positions of influence at the English court were already occupied by the English nobility.

Obstreperous and independent‐minded as they tended to be, the Anglo‐Norman barons in Wales and Ireland nevertheless were supporters of the English crown at least in the sense that they were not going to repudiate it. The English crown was their crown, and with their help a degree of English control or at least influence could be maintained in parts of Ireland and Wales (even if in the pre‐Reformation period it always remained tenuous). What made Scotland different was that there the same group had an alternative kingship to adhere to, which they did. The character of Edward I, aloof, severe, ungenerous, must have helped to push them in that direction, but this was a contributing factor in a development for which the existence of an established Scottish kingship was a necessary condition.

Scottish kingship was imperfect in that incumbents had hitherto not been crowned and anointed, but that deficiency was remedied in due course. Here, the attitude of the pope was crucial. The declaration of Arbroath was addressed to the pope; but Scottish lobbying at the curia was intense already in the preceding decades, and quite successful. Following the Scottish revolt of 1297, pope Boniface VIII sent Edward a stern missive in which he claimed that the kingdom of Scotland was a papal fief, not an English fief, and demanded of the English king to justify his behaviour regarding Scotland by a set date (papal bull Scimus fili, 1299). In fact, unlike the kingdom of Sicily, or Ireland (where this was derived from the fake Donation of Constantine), Scotland most certainly was not under papal suzerainty. This papal claim may have been a calculated gamble, or an (p. 386 ) honest mistake perhaps based on the papal bull Cum universi of 1189 or thereabouts, in which, at the request of the Scottish clergy, the Scottish church had been exempted from the ecclesiastical jurisdiction of the English archbishop of York and placed directly under papal jurisdiction. ⁹³

Papal pressure seems to have been responsible for the liberation of John Balliol. He withdrew to the old Balliol family seat in Picardy but seemed poised to be reinstated as Scottish king. This prompted the Scottish rebels to date their official utterances by the years of his reign from 1301 onwards, and Robert Bruce junior to leave their ranks and defect to the English again. When, despite Robert Bruce the elder and his son of the same name having fought for Edward against John Balliol, the Bruce family nevertheless did not receive the Scottish kingship, Robert Bruce junior joined the Scottish insurgency and indeed became a member of the regency council. Switching sides again, in 1302 he made his peace with Edward, a move no doubt dictated by several motives. Robert had no gratitude to expect from John Balliol; he may have hoped that Edward would now back his candidacy for the Scottish throne after all, as an alternative to the humiliating Balliol restoration; and his father Robert Bruce senior might not live much longer (indeed he died in 1304): as one of the Scottish rebels, Robert junior would not have been able to succeed to the family estates in England.

The difficulty with which his Scottish policy had met may have contributed to Edward's decision in 1301 to create a new Prince of Wales: his son, the future Edward II. Indeed that young prince soon achieved considerable popularity in Wales, another indication that ethno‐cultural nationalism was not the chief root of Welsh political discontent. In the face of domestic English opposition to the exactions imposed by the crown to finance its military efforts, and of the Scottish rebellion, Edward abandoned the French war in 1297. He concluded a truce with king Philip but continued to fund adversaries of his French opponent in the Low Countries and the Rhineland. This strategy scored a major success when, in 1302, Philip suffered a disastrous defeat at Kortrijk at the hands of the Flemings (encouraged by Edward). Unfortunately for the Scots, this forced Philip to convert his truce with Edward into a full‐scale peace and restore Aquitaine. Already in 1299, Edward had married Philip's sister; now the Anglo‐French reconciliation was cemented by the engagement of the English crown prince—the new Prince of Wales, the product of an earlier marriage by Edward—with Philip's daughter Isabella.

Also in 1302, pope Boniface issued his bull Unam sanctam, which destroyed what little goodwill was left in the already strained relationship between the French court and the curia. With the French king now working to have the pope deposed, Boniface naturally grew more sympathetic to Philip's rival Edward, and accused the Scottish episcopate (quite correctly) of fomenting the Scottish insurgency. In this situation the pope became inclined to accept Edward's reply to Scimus fili, to the effect that when Edward's forbears the Trojans (led by Aeneas’ son Brutus) had conquered Britain (named, of course, after Brutus), they had also subjected the Scots—even if the Scots countered immediately with a story similar to that later included in the declaration of Arbroath, that they, invaders themselves, had come to Scotland via Egypt and Spain and had never been subject to anyone. ⁹⁴

(p. 387 ) Following another successful campaign, Edward imposed a peace on Scotland in 1304 that most of the Scottish magnates accepted. Meanwhile, Robert Bruce junior was once again disappointed with Edward's failure to reward his (new‐found) loyalty to the English crown, indeed to treat him with any regard. Of course Robert had proved fickle; the English king may have felt that pardoning Robert was already more than he deserved, to say nothing of giving the man a throne to sit on. Anyway, with king Philip and pope Boniface no longer pushing for the restoration of John Balliol this had ceased to be a pressing concern. But Robert was a major Scottish player with a good dynastic claim to the Scottish kingship, and in retrospect it is clear that the English king made a mistake in not trying at least to string him along.

There was a third great Anglo‐Norman family in Scotland with a claim to the throne, the Comyn family. To get anywhere in Scotland, Robert needed their support, or at least their neutrality. But he trusted them so little (or they him, or both) that when they arranged to meet it was in the assumed safety of a sanctuary, the greyfriars’ church at Dumfries (1306). To no avail: the exact circumstances are unclear, but at the end of the meeting the senior Comyn was dead and Robert a fugitive guilty of so major a sin—killing someone in a church—that it made him automatically excommunicate. To be sure, the bishop of Glasgow helpfully granted him absolution straightaway, but the curia would not hear of it. Reckoning no doubt that in the face of this disaster nothing could conceivably make things worse, Robert hastily had himself inaugurated king of Scots at Scone by a handful of loyal adherents, in a makeshift ceremony and with an ersatz sacred stone. His sole chance of turning the situation around and saving himself from total ignominy was to get accepted as Scottish king. The odds were long: he was up against king Edward, the Comyn family and their following, those in Scotland who thought that their rightful king was John Balliol, and the pope. But, his ambition fuelled further by the absence of any alternative, Robert succeeded. Hunted by Edward, who imprisoned his wife, daughter, and other female relatives and had three of his brothers executed, he made himself the soul of the resistance, outshining the passive John Balliol. John died at Bailleul in 1313; already in 1309, at a parliament at Saint Andrews, the Scots had transferred their allegiance to Robert.

Luck was on his side in that the English king did not have much time left to deal with him. In 1307, Edward was succeeded by the incapable Edward II, who alienated the English nobility by heaping extravagant favours on his equally incapable lover Piers Gaveston. Although Edward II remained committed to crushing the Scottish resistance, he lacked his father's single‐minded resolve; moreover, the walking domestic crisis that he represented distracted from the war effort. In 1314 at Bannockburn, Robert inflicted a crushing defeat on the English army. In 1316, the pope readmitted him to the church, only to excommunicate him again in 1318 in the context of a quarrel over the nomination of Scottish bishops. It was in this context that the declaration of Arbroath was dispatched in 1320; in 1324, the pope recognized Robert as king. Caught up in domestic turmoil, Edward II meanwhile pursued the Scottish war only languidly.

In 1327 he was deposed by his French wife, Isabella, who assumed the regency for her son Edward III. A new campaign against the Scots having failed, the cash‐strapped queen made peace with Robert in return for a payment of 20,000 pounds. The peace involved the marriage of her daughter with the Scottish crown prince, David Bruce. In a document negotiated at Edinburgh in 1327 and ratified at Northampton in 1328, Edward III relinquished any claim to suzerainty over Scotland. Robert now dispatched another (p. 388 ) petition to the Holy See asking for permission to be crowned and anointed. This time it was granted: Robert having died in 1329, his little son David II in 1331 became the first king of Scots to be christus Domini, as, likewise, were his successors. Although Isabella had agreed to return the stone of Scone, it was not delivered to her by its keeper, and she travelled north to the marriage of her daughter at Berwick on the English–Scottish border without it. ⁹⁵ It seems that the Scots did not make a fuss, either then or thereafter: the reason for that may well be that the stone had become dispensable.

Thanks to the acquisition of a proper crown the kingdom of Scotland was now indestructible, even though its relations with its southern neighbour remained turbulent and the power of the Scottish crown, negligible. Indeed, in the fourteenth century the country basically sank into anarchy. Edward III of England seized power from his mother in 1330 and repudiated the ‘quitclaim’ of 1328 that freed the Scottish king from any obligation to the English crown—on the grounds that although it was issued in his name he had had no part in it. The new English king supported the claim of Edward Balliol, son of John, to the Scottish throne. In return for English military backing, Edward Balliol recognized English suzerainty, landed in Scotland, and had himself inaugurated at Scone (I have been unable to find any reliable information about whether this involved a coronation). While Scotland once again came under English occupation, young king David was sent to the safety of France, only to be captured by the English at his very first military venture against them. A nominal prisoner at the English court, he felt so at home there that he returned repeatedly even when, eleven years later, he was released for a huge ransom raised by the Scots; moreover, it was with difficulty that, on at least two occasions, the Scots prevented him from appointing the English king (his brother‐in‐law) as his heir if he died childless, which he did.

The Scots meanwhile got rid of Edward Balliol, who despite military victories by his English troops was unable to establish any real control over the country. With king David in exile, the realm was administered by his nephew, who also succeeded him at his death in 1371; his original position as steward of Scotland gave the new dynasty its name (albeit usually in the spelling Stuart). Hampered by power struggles within the dynasty, long periods of regency (in the fifteenth century, king after king died young, leaving under‐age heirs), powerful magnates, and intermittent warfare against the English (with, at best, inconclusive results), Scottish kingship after the reassertion of independence in the early fourteenth century never gained much strength. But, significantly, it was never again contested by the English crown. Its independence was not the result of its ‘deterrent’ capability (except perhaps in the sense that the English had learned the lesson that it was unwise to stir up this hornets’ nest). Rather, by the late pre‐Reformation period destruction of an established kingdom had become simply unthinkable.

In the ancien régime, a kingdom might—occasionally, and typically to a quite limited extent—expand through conquest. But the only way to join another kingdom, as such, to one's own was to marry or inherit it. Merger by conquest was simply out of the question—unless you went to war in pursuit of a title based on heredity. Not infrequently, more than one claimant might have such a title, and it might be genuinely impossible to determine which of them had the best. In such a situation, war was an accepted way of settling the question. The attempt by Edward Balliol to displace David II is a case (p. 389 ) in point: his dynastic right to the Scottish throne was as good as David's. Scotland and England eventually were joined, in the manner typical of the ancien régime, and entirely peacefully: the Stuart dynasty inherited the English throne when the Tudor dynasty became extinct in 1603.

4.2.3.4 Ireland and Wales II: absent a native crown, English overlordship proves impossible to remove

If the Bruce family successfully claimed and consolidated the Scottish kingship (because it was there to claim in the first place), they failed in Ireland, where they tried, too. After the usurpation of 1306, Edward I of England had all the Bruce brothers killed but two—Robert himself and Edward. When Robert was recognized as Scottish king at the parliament of Saint Andrews in 1309, he had no son, which made Edward his heir‐presumptive. But once Robert was reunited with his wife in 1315, as a result of a prisoner exchange following his victory at Bannockburn, it was reasonable to expect that he would in due course produce a direct heir. A plan was conceived to find Edward, who seems to have been a difficult and unruly character, a kingdom of his own in Ireland. For Robert, this had the double advantage of getting Edward out of his hair while opening a new front for the English in Ireland, diminishing the resources that they could direct against Scotland. In 1315, Edward thus landed in Ireland and had himself inaugurated as king of Ireland at Dundalk. However, his reception in Ireland was far from enthusiastic, not just among the Anglo‐Irish, but also among the ‘Irish’ Irish. The Scottish attempt to take Dublin by surprise having failed, in 1316 Robert Bruce himself landed in Ireland to aid Edward militarily, but withdrew without achieving any breakthrough; thereafter military operations continued inconclusively until Edward was killed in battle in 1318. Already in the previous year, an Irish relative of the Bruce brothers, Domnall Uí Néill (Donald O'Neill), had sent the pope an address in support of the Bruce claim to the ‘Irish’ throne, a document that is commonly if misleadingly referred to as the ‘Remonstrance of the Irish princes’. ⁹⁶

It has been compared with the declaration of Arbroath and even, incongruously, attributed to the same unknown drafter. In fact, it conspicuously lacks the rhetorical and psychological skill apparent in the Arbroath document. The 1317 remonstrance is much longer, indeed excessively so, much of it taken up by an endless and, in part, intrinsically implausible rant against the misdeeds of the ‘English’ (including the Anglo‐Irish) in Ireland. The remonstrance recognizes the English claim to Ireland based on Laudabiliter, but argues that because the English had failed to improve Irish morals (or levy the tax for the Holy See provided for in the bull) they had forfeited that claim. At the end of the document the pope is informed, rather summarily, that because the English had not carried out the mandate given to them in Laudabiliter, and moreover were impossible to live with, the Irish had chosen Edward Bruce to be their king instead of the English king. Thus, having recognized papal suzerainty over Ireland the document goes on to ignore it, without any apparent awareness of the contradiction. By contrast, the declaration of Arbroath cleverly combines pressure on the pope (who is warned that any further bloodshed in the English–Scottish conflict will imperil the salvation of his soul unless he leans on the English king to prevent it) with promises (do what we want (p. 390 ) and we shall go on a crusade), and a great rhetorical display of humility and deference. And whereas the remonstrance assails the perfidy of the English at inordinate length and returns to this topic at every opportunity, the declaration, having gravely but with adroit concision alluded to English cruelty, continues in an inclusive and conciliatory register (no difference, for christians, between ‘Jews and Greeks, Scots and English’; Scottish readiness ‘to do for [the English king] what is in our power’).

The main similarity between the remonstrance and the declaration (it is of course possible that, even if they were not composed by the same person, whoever drew up the declaration knew the earlier remonstrance) is the emphasis on the ancientness of the native kingship: the Arbroath document advertises 113 native Scottish kings, the line unbroken by any foreigner; the 1317 remonstrance declares Domnall to be heir to 197 native Irish kings, and to ‘the whole of Ireland’. But although the Ui Néill had indeed frequently held the post of ard rí, this was no ‘Irish kingship’ comparable to the Scottish kingship. In the document Domnall actually describes himself as rex Ultoniae, ‘king of Ulster’, without explaining how the existence of a separate Ulster ‘kingship’ squared with the ‘Irish’ kingship that he postulated: it did in the old Irish system outlined above, but this could of course not be invoked in support of the kind of kingship Domnall was trying to obtain for Edward Bruce. Nor does Domnall explain why, if he himself was heir to a long line of native kings, that kingship should now be given to someone else, and a foreigner, too. In a brief throwaway phrase the text does assert Edward to have 'sprung from our noblest ancestors’, but even though the Anglo‐Norman nobility in Scotland had intermarried with the native nobility, and Edward was related to Domnall, he could hardly be described as Irish. The Bruce family, ironically, was just as Anglo‐Norman as the Anglo‐Irish barons that the remonstrance complains about—indeed, as mentioned, Robert Bruce was the son‐in‐law of the Anglo‐Irish earl of Ulster. (Domnall was actually rí of Tir Eógain/Tyrone, and in the sixteenth century an ‘earldom of Tyrone’ was created for the Ui Néill.)

The declaration of Arbroath was issued by some forty Scottish magnates that it lists by name and whose seals (or, at any rate, a corresponding number of seals: some of them do not match) were appended. By contrast, the remonstrance, even though it purports to be speaking for the Irish as a whole, only lists Domnall by name. Moreover, the declaration of Arbroath was dispatched after a successful Scottish revolt against the English invaders; but the 1317 document announces an Irish revolt that was yet to take place. It never did—even though, if the Irish had wanted to stage a rebellion, this would have been the moment to do it. Their failure to do so, along with many of the oddities of the text, is explained if rather than taking the document at face value the whole undertaking is interpreted as very much a Scottish initiative with minimal Irish input—which of course Domnall could not let on to. Robert Bruce had hoped that after Bannockburn the English king would be ready to negotiate, but when that expectation turned out to be wrong Robert must have been eager to increase the pressure on the English further. In fact they came to believe (possibly encouraged by Robert himself) that he would land in Wales and launch a rising there, so that the English were preparing to fend off an invasion of Wales when what occurred was an invasion of Ireland. As a result, English forces were tied down in Wales and Ireland. Stuck in Ireland, Edward in 1316 issued an appeal to the Welsh magnates to make him the leader of a revolt against the English—evidently falling back on plan B; no rising, however, occurred. Nevertheless, northern England, militarily left to its own devices (p. 391 ) throughout most of the remainder of Edward II's reign, lay open to devastating Scottish raids.

As Archibald Duncan has ingeniously and convincingly shown, the common assumption that Edward Bruce adopted the Irish kingship only in 1316, over a year after his arrival on the island, is due to a chronicler's error. In fact, the inauguration at Dundalk must have taken place very soon after Edward had disembarked—presenting the Irish with a fait accompli rather than taking place at their invitation. On the contrary, it was an invitation to them—but which they largely failed to warm to. Nor, it would appear, was the bulk of the 1317 remonstrance—the jeremiad against English mistreatment of the natives—originally connected with the Bruce mission, to which reference is made only at the end of the document. A pre‐existing text seems to have been adapted for the purpose of providing a semblance of legitimacy for Edward. ⁹⁷

In Wales, the single great revolt against English domination took place over a century after the integration of the country into the English kingdom. Interestingly, in light of the thesis proposed here that after the turn of the millennium independence of a country rested on the presence of a crown, there seems to be no evidence that the leader of the revolt, Owain Glyn Dŵr (Owen Glendower), ever explicitly sought independence for Wales or a crown for himself. Such a new crown would by this time have been very difficult to create, against the opposition of the English crown and given that English suzerainty had been acknowledged for so long and by so many, including Owain himself.

Owain took advantage of the revolt of leading English nobles against the English king: the incumbent, Henry IV Lancaster (a branch of the Plantagenet dynasty), being of doubtful legitimacy since he owed the throne to the deposition of Richard II in 1399, the rebels wanted to replace him with another member of the dynasty (an ancestor of what would later become known as the York branch of the royal house). Owain was a well‐to‐do member of the Welsh gentry, linked to the English court by his career and marriage—and thus no obvious victim of any ‘colonial’ oppression. By the terms of the so‐called Tripartite Indenture concluded by the rebels in 1405, Owain was to become prince of a Greater Wales, territorially augmented by several English counties adjoining the principality. A copy of a treaty that Owain concluded with the French king in 1404 and now in the British Public Record Office preserves his magnificent seal. It shows him sitting on a throne, behind which angels hold up his mantle, embroidered with heraldic emblems—the lions of Gwynedd, from whose rulers he claimed to be descended though apparently he was not (he was related to the ancient rulers of Powys and Deheubarth). The impression is certainly regal. But although the figure on the throne is holding a sceptre, his head is bare.

The legend reads owynus dei gracia princeps wallie, Owain, by god's grace Prince of Wales. Kings would routinely date charters as having been issued ‘in the so‐and‐so‐manieth year of our reign [regni nostri]’; Owain put ‘in the so‐and‐so‐manieth year of our being prince [principatus nostri]’. The ambiguity of his position and aims was never resolved. In the feudal hierarchy of christendom, a mere ‘prince’ needed a (royal) liege. Presumably, if Owain had prevailed he would nevertheless have done homage to the English king, especially as it is hard to see how he could have got away with establishing an independent Welsh kingdom that would actually have incorporated a substantial amount of unquestionably English territory. In any case he did not prevail, because (p. 392 ) Henry IV defeated his English adversaries and held on to his throne. For some time, Owain's cause was quite popular in Wales, feeding on mutual animosity between the Welsh and the English. But in the end, the rising subsided as abruptly as it had begun; the self‐appointed prince simply vanished, and we do not even know when or how he died (probably in 1415 or 1416). Many Welsh magnates no doubt preferred the English crown to a shady Welsh dignity whose precise nature was yet to become clear, and which had no real roots in Welsh tradition since the country had never been united under one of its native leaders—except briefly and still incompletely by Llywelyn ap Grufudd, who at the time was actually resented by many Welsh. ⁹⁸

Probably quite typical is the case of one Welsh family descended from a court official of Llywelyn ap Grufudd and as yet without a surname, in place of which Welsh tradition used the ap 'son of’. In the fifteenth century, rather than changing in every generation, the name following the ap was often adopted as a fixed patronym. In this manner one Owain, son of Maredudd ap Tewdwr (Meredith son of Theodore), used his grandfather's name as a surname and called himself Owen Tudor in English. Following the occupation of Wales by Edward I and the death of Llywelyn in 1282, some members of the family had made peace with the English king immediately. Two direct forbears of Owen Tudor, Tewdwr Hen (i.e. the elder) and his son Goronwy, joined the rebellion of 1294, but they, too, ultimately made peace with Edward and served in the royal administration of north Wales. When Owain Glyn Dŵr rebelled, the family dominated Anglesey. Kinsmen of Owain Glyn Dŵr, two uncles of Owen Tudor joined the rebellion. Owen himself, too young to have any part in it (he was born around 1400), apparently profited from efforts by Henry V to conciliate the Welsh after the collapse of the rising, and gravitated to the royal court. Despite being of no particular prominence, he managed in 1429 secretly to wed the widow of king Henry VI, Catherine of Valois. The York and Lancaster branches of the royal house having fought over the crown until they were both extinct, in 1485 his descendant Henry Tudor mounted the English throne (champion of the already extinct Lancaster branch, he defeated the last York claimant, killed in that battle).

4.2.3.5 France and the Hundred Years War

As in the case of Scotland, the expansionist English kingship was unable to gain control of the kingdom of France, despite military superiority and a presentable claim to the French throne. In 1328, the direct male line of the Capet dynasty became extinct, as the sons of Philip IV, who succeeded him, left no male heir. This left king Edward III of England (reigns 1327–77), the son of Philip's daughter Isabella, as Philip's sole male descendant. Women were barred from succeeding to the French throne, and so there was no question of Isabella ascending the throne herself. But it was not clear whether women could pass on the right of succession. If so, then Edward III of England after 1328 was also king of France by right of inheritance. If, on the other hand, the exclusion of women from the throne was taken to imply that not only the crown, but also the right of succession could only be passed on in the male line, then the heir to the throne was Philip's nephew, the count of Valois (son of Philip's brother Charles, the unsuccessful rival of Henry of Luxemburg in the election for the Roman–German throne of 1308). (p. 393 ) Either interpretation could be argued with some plausibility—the typical backdrop of a war of succession in ancien régime Europe. In France, the prospect of the French crown being united to the English crown was not welcome. It would have meant depriving the French nobility of a source of prestige and preferment with which they had established ties in favour of magnates whose main attachment was to the English court. This, more than any ‘national’ sentiment, is likely to have been the main reason why the French preferred the Valois succession. After all, the English crown was not culturally alien. The English king was duke of Aquitaine, that is, one of the greatest lords in the French realm, and French was the language of the English court and the English nobility.

But Edward could not be expected simply to give up his claim. Militarily, he was in a promising position. His French possessions gave him a bridgehead on the continent, while the fact that England was an island made it difficult to strike at his base. What, in spite of such resounding victories as Crécy and Azincourt (the correct spelling), led to the ultimate defeat of the English crown in the ‘Hundred Years War’ was simply that the French, like the Scots, preferred to keep a king of their own. Edward I had sought to tread a fine line between confiscating Scottish estates, awarded to English magnates as an incentive for assisting him in the durable subjugation of Scotland, and not confiscating Scottish estates so as to keep some backing among the native magnates. Edward III resumed the attempt to bring Scotland under English control (through the client kingship of Edward Balliol) in part to allay the dissatisfaction of English nobles deprived of their Scottish gains by the earlier recognition of the Bruce kingship. What in Scotland proved unfeasible would have been even more difficult in France. To get military support in England for its French war, the English crown needed French spoils. But that made it impossible to get the political support of the French nobility—which was at least as necessary for the English crown to win recognition of its authority in France.

Edward III may have realized this. It is unlikely that he started the French war with the intention actually to deprive his Valois cousin of the French throne. Rather, he must have hoped to be able to sell his dynastic claim—for doing which it was necessary to assert it militarily first, to show his determination not to give it up for free. The most obvious prize that he could aim for was a quitclaim for Aquitaine, removing the irksome obligation on the English king to do homage for this fief to the French crown, and instead incorporating it in the English kingdom proper. From a French perspective, however, this was already a huge concession, not to be granted except in an extremity.

Edward successfully created that extremity, defeating and capturing king John II of France at Poitiers in 1356. He even laid siege to Reims in 1360—as if threatening actually to have himself crowned king of France, a ceremony normally held in Reims cathedral. The fact that the siege was quickly lifted suggests, however, that it was essentially symbolical, a means of putting pressure on the French. Indeed they agreed on a sweeping peace settlement, involving the cession of Aquitaine in return for Edward abandoning his claim to the French throne. But the settlement was eventually ratified by neither side—ultimately no doubt because it was so unbalanced in favour of the English that even they did not consider it viable in the long term. King John was released in return for a promised payment of three million écus to the English crown. Hostages at the English court were to vouch for this, among them a younger brother of the French crown prince. When he returned to France without authorization, king John voluntarily went back to London, where he died in 1364—a striking demonstration of attachment to a chivalrous code of honour.

(p. 394 ) His successor Charles V avoided major showdowns with the English, which the French had an alarming tendency to lose, and preferred diplomatic means to check the ambition of the English king. Count Louis of Flanders wanted to marry his daughter and sole heir to a younger son of Edward III, Edmund. This would have given the English dynasty another of the great French principalities in addition to Aquitaine. King Charles, for his part, sought the hand of the Flemish heiress for his brother Philip, whom he had created duke of Burgundy. Both Edmund and Philip were so closely related to their prospective bride that the marriage required a papal dispensation. It was Charles who secured it, not Edward, a triumph for the French king—even if it ultimately led to another big problem for the French monarchy in the shape of the virtually independent ‘Burgundian’ realm, a development that at the time could not be foreseen. The episode again illustrates the huge importance of legal considerations for the politics of the period, where warfare was unlikely to be successful unless it was in support of what could plausibly be considered a valid legal claim, or if at least it did not run too visibly counter to any such claim. No amount of English military strength could alter the papal decision awarding Flanders to the Valois rather than the Plantagenet dynasty.

In subsequent generations the (rather intermittently fought) war between the two royal houses became more acrimonious. Unable to turn their military prowess into any lasting settlement in their favour, the English grew more reckless. This culminated in the coronation of the English boy king, Henry VI, as king of France in Paris in 1431, after king Charles VI of France had been prevailed on to marry his daughter Catherine to Henry (she was later to marry Owen Tudor) and disinherit his son, the future Charles VII, in favour of the English crown. But the timid Charles was famously persuaded by Joan of Arc to have himself crowned at Reims anyway. Reims was the right place to do it, not Paris, and this symbolized the greater legitimacy personified by Charles. The English side had gone too far, and in the end the English crown (which stubbornly continued to claim the title ‘king/queen of France’ throughout the ancien régime) even lost Aquitaine.

We saw that both Thomas and Giles treat the king as legislator. The concept as such they found in Aristotle, but according to Roman law in its Justinianic redaction this was a capacity reserved for the Roman emperor. On the other hand, it was obvious that the emperor had long ceased to control christendom in its entirety, resulting, by the turn of the thirteenth century, in a current in legal thought that tended to negate the difference between kingship and the imperial office. Thus, in the 1190s the English legal scholar Richard de Lacy noted that ‘both emperor and king owe their office to [literally: are anointed by] the same authority [the church], the same consecration, the same chrism. Where, therefore, would any difference in their powers come from?’ ¹¹⁴

That there was no such difference meant that rex est imperator in regno suo, a king is emperor in his own kingdom, in the famous formula coined around the same time by Italian lawyers. ¹¹⁵ Pope Innocent III, himself a legal scholar, gave added weight to this position in his decretal (legal opinion) Per venerabilem of 1202, which famously says of the French king that ‘in temporal affairs he recognizes no superior at all [superiorem in temporalibus minime recognosc(i)t].’ Therefore, according to Innocent, the French king had the right to turn directly to the pope to have an heir issued from a marriage of doubtful validity confirmed by the pope, whereas a prince subject to a feudal superior, like count William of Montpellier, to whom the document is addressed, did not. The position here adopted by Innocent was a double slight to the Roman crown: not only did the pope implicitly deny its superordination to the French crown, but the legitimation of bastards was normally held to be an imperial right. Innocent is thus in effect annexing part of the imperial office to the papacy (on grounds set out at some length in the decretal), in line with a trend towards turning the pope into some sort of emperor himself. Indeed, although saying that the French crown had no temporal superior sounds magnanimous, both the arrogation of the right to pronounce on this and the exercise, vis‐à‐vis the French king, of the imperial right to legitimize bastards went some way towards asserting papal suzerainty over the French crown. In another decretal of the same year, Venerabilem fratrem, Innocent explicitly claimed suzerainty over the Roman crown, declaring that the newly elected king of Romans needed to be approved by him (a position which, however, was never accepted in Germany); and in 1213 Innocent made king John of England do homage to him and submit to papal suzerainty for his kingdom in order to be received back into the church. This trend towards an imperial papacy culminated in, and failed with, Boniface VIII. We saw how contemporary chronicles depict Boniface with imperial insignia and even claiming the imperial dignity for himself as he rejects the request of the king of Romans, Albert, to be crowned emperor in 1298. ¹¹⁶

His imperial pretensions, in particular the bull Unam sanctam of 1302, for once led to a massive falling out between the pope and the French king rather than the Roman crown, and indeed to a rapprochement between Boniface and Albert. Reversing the stance taken by his predecessor Innocent a hundred years earlier, the pope peremptorily declared the French king to be subject to the Roman crown (Albert in turn had declared himself (p. 404 ) homo papae, accepting papal suzerainty). ¹¹⁷ This interlude was shortlived: ten years later the emperor Henry VII once more found pope Clement V and the French king united (more or less) in their opposition to him, reaffirming the prevailing pattern of conflict between the papacy and the Roman crown. Not least simply because the territory of the latter included parts of Italy, it was tempting for the papacy to play more distant rulers against it—more particularly the French king, who was happy to go along with this. (The opposition of the papacy to the emergence of a rival power centre in Italy was a recurrent pattern ever since it invited the Frankish king to attack the Lombards in the eighth century, the very move that, ironically, led to the re‐establishment of the western imperial dignity in the first place.) Had they been consistently united, pope and emperor might have been invincible; as it was, their disunity weakened both and helped promote the autonomy (never actually threatened) of the other realms.

From 1250 to 1312, there was no emperor, and the king of Romans was weak. In such circumstances it made little sense to deny that a king could make laws just as the emperor once did. Although some still had misgivings about this, Thomas and Giles already take it for granted. Indeed their (utopian) king is clearly an absolute ruler, like the Roman princeps of old—or the pope of their own day: not fortuitously, Giles of Rome at a later stage of his career became an outspoken supporter of the notion of papal suzerainty over christendom, including in secular matters, a position expressed in his other famous treatise De potestate ecclesiastica (‘On the powers of the church’) of 1301/2. As we saw, Thomas and Giles in their writings on kingship do not discuss the issue of political organization beyond the level of the kingdom. But by the fifteenth century, even those who developed a theory of the empire—concretely, Nicholas of Kues and Enea Silvio Piccolomini—expound a notion of rulership that is closer to the activist kingship portrayed by Thomas and Giles than to the conservative, minimalist conception of the imperial office put forward by Engelbert and Dante. Both Nicholas and Enea Silvio were travelling further in the direction indicated earlier by Thomas and Giles, along a road that was to lead to the ‘absolutism’ of the post‐Reformation period.

4.2.5.1 Nicholas of Kues and Sigismund of Luxemburg

In the fifteenth century, the notion of Latin christendom as a single community repeatedly found dramatic expression in the form of general councils, assemblies not just of church dignitaries, but also of secular notables from all parts of christendom—in particular the councils of Konstanz (1414–18) and Basel (later moved to Lausanne, 1431–49). The councils were a forum for discussing the reform of the church, the current condition of which was the object of widespread and fundamental criticism. Foreshadowing the Reformation, already the Hussite movement in Bohemia had separated from the church of Rome in an attempt at radical reform, providing another topic of debate by the councils. The councils were dominated by a grass‐roots movement calling for what one is tempted to call a democratization of the church, and which in the historiographical literature goes under the name ‘the conciliar movement’. One major focus of criticism was the role of the pope, whom many wanted to put under the control of the councils—in a manner comparable to the efforts of the assemblies of estates that in the fifteenth century also sprang up or were consolidated in the secular realms and principalities of christendom, (p. 405 ) and which aimed at establishing a degree of control (in this instance, financial control in particular) over their princes.

Among the tasks facing the council of Konstanz was ending the Great Schism, the rivalry between competing popes that had been going on for decades. The council therefore declared that it ranked higher than the popes (in the decree Haec sancta, 1415), initiated formal legal proceedings against two of the three claimants to the succession of Peter at this time, and, having condemned them, declared their deposition, this being a precondition for the voluntary withdrawal of the third candidate. The council then elected a new supreme pontiff to replace them all. The work of the council was facilitated by the king of Romans—as well as of Hungary—Sigismund of Luxemburg (great‐grandson of Henry VII; nominally, in 1419 he also succeeded to the throne of Bohemia but until 1436 was prevented by the Hussites from actually occupying it). His arrival at the council in 1415 was the occasion for a great pageant, but his role in the proceedings was not merely symbolical. In an attempt to resolve some of the major quarrels within christendom, in the next two years he undertook a great deal of travelling—for example to the Aragonese court at Perpignan (where he held negotiations that succeeded in prevailing on the Spanish rulers and the king of Scots to withdraw their support for ‘their’ candidate to the papacy), and then to Paris and London (where he unsuccessfully tried to end the Hundred Years War). In 1417 Sigismund held a second, and even more spectacular formal entry into Konstanz. ¹¹⁸

The papacy unsurprisingly took a detached view of the ‘conciliar movement’ and its activities. The council of Konstanz having, in the decree Frequens of 1417, mandated the calling of a new council after five years (then after seven years, and then once every ten years), pope Martin V—the one elected at Konstanz—reluctantly convened one at Pavia in 1423. It was poorly attended and soon dissolved. Sigismund was busy fighting the Turkish threat to his other kingdom, Hungary, and unable to intervene effectively. The next council assembled at Basel in 1431; like its predecessor it was at first attended poorly. The new pope Eugene IV was in a hurry to disband it again, but the council for its part initiated legal proceedings against him on the grounds that he had been elected in contravention of canon law. This generated much attention for it and attracted a big crowd of participants (anyone in christendom with either a title of nobility or an academic degree was free to take part—he simply had to register and swear an oath to observe the regulations governing the proceedings). Sigismund was on his way to Rome to receive the imperial crown when news reached him that Eugene had dissolved the council and convened another in Bologna. Inviting the Basel council to stay in session, Sigismund at the same time called on pope Eugene to rescind its dissolution in return for support against the accusations levelled at him by the council. Eugene complied, and Sigismund in 1433 held another triumphal entry at a general council of christendom, this time as freshly crowned emperor. He died in 1437. In 1439, the council declared Eugene deposed after all and elected an anti‐pope. The new king of Romans Frederic of Habsburg (reigns 1440–93) at first remained neutral, but then inclined towards Eugene. Thus, against the background of the quarrel between the conciliar movement and the papacy, the emperor (whether or not he had already formally received that title—both Sigismund and Frederic eventually did) once more played an important role in christendom, or at the very least enjoyed a high degree of visibility.

(p. 406 ) Among the participants of the Basel council was the theologian and philosopher Nicholas of Kues (a town in the Mosel valley, near Trier; the Latin version of the name is Nicolaus Cusanus or de Cusa) (1401–64). In 1433—the emperor Sigismund had just arrived in Basel—Nicholas addressed to the council and to the emperor a long treatise on no less a subject than christendom as a whole, known under the title De concordantia catholica (‘On the universal concordance’ or ‘On the concordance of all things’). The treatise deals with the church in the wider sense (the christian people, but also supernatural beings like the different categories of angels, or god himself), the church as an earthly institution (whose hierarchical character, culminating in the pope, Nicholas underlines), and the secular organization of christendom. This is the subject of the last of the three books into which the treatise is divided, and takes up more than half of it. Significantly, for Nicholas the question of the secular organization of christendom is still simply identical with the question of the nature of the ‘empire’. Nicholas demonstrates how different the perception of political ‘reality’ (to the extent that such a thing exists) at that time could be from what we are used to nowadays. An age for whose most astute intellects the existence, for example, of the nine choirs of angels was an unchallengeable given treated the question of the reality of the Roman empire in a different way from what would come naturally to us. ¹¹⁹

In the introduction to Book III (§§268–91), Nicholas sets out the basic tenets of his political thinking, garnishing them with copious references to Aristotle—references gleaned from the Defensor pacis of Marsilius of Padua of 1324, a source that Nicholas does not acknowledge but the mistakes of whose quotes from Aristotle he faithfully reproduces; much else is borrowed from Marsilius, very much a proto‐conciliarist thinker, too. ¹²⁰ According to Nicholas, correct organization of any human community is predicated on rule over that community being exercised by the consent of the ruled; on the other hand, rule is indispensable because of the unequal intelligence of men and their unequal ability to hold evil passions in check. It is a requirement of nature that the wise (sapientes) should lead those who lack wisdom (insipientes), with the voluntary subjection (voluntaria subiectio) of the less wise to their betters dictated by a natural instinct (naturalis instinctus) and resulting in ‘a kind of natural servitude [quaedam naturalis servitus]’. ¹²¹

Among the various forms of rule, rule by a single person was preferable (§§282–3), though that person had to respect the laws where they existed. In this context, Nicholas quotes Aristotle to the effect that in the politeía, the proper, legitimate political community, only the laws should rule. ¹²² For Nicholas, the passing of laws (legis latio) was not the task of the ruler, but of all those who would be bound by the laws in question, or at least their elected representatives. ¹²³ The ruler himself should be elected, if possible without regard for any hereditary claim, with the participation ‘of all, or the majority, (p. 407 ) or at least those among them of highest rank’. ¹²⁴ This sort of thing is quite in line with ‘conciliarism’; indeed, Nicholas also expressly endorsed the view that the authority of the council was above that of the pope. ¹²⁵ It is likewise in line with (and evidently to a large extent copied from) Marsilius. Marsilius, too, had already stressed the authority of the council and assigned it prerogatives—for example, the right to elevate someone to sainthood—that were in fact exercised by the pope; he had also ascribed to the council authority to depose the pope.

At the same time, however, Nicholas was in favour of a markedly hierarchical social organization, with a strong position for the ruler. This was not very ‘conciliar’, and it seems that among the Basel radicals Nicholas felt increasingly uneasy. In 1437, he left Basel to join the counter‐council that Eugene had convened in Ferrara and which was later moved to Florence. Eugene subsequently made Nicholas bishop of Brixen and created him a cardinal. In his later writings Nicholas keeps away from political subjects, and when he himself produced an edition of his collected works, the Concordantia catholica was not included. Perhaps the hastily written treatise by then seemed half‐baked to him; besides, he may have been embarrassed by his flirtation with ‘conciliarism’.

After the general introduction to Book III of the Concordantia, inspired by Marsilius, Nicholas proceeds to his discussion of the empire. For him—as a necessary corollary of the hierarchical organization (hierarchica ordinatio) of society, concretely christendom—the king of Romans is the first among kings, just as the bishop of Rome is the first among bishops. ¹²⁶

In what way is the ‘holy empire’ ‘of god’ (quomodo imperium sacrum a deo)? (For the expression ‘of god’ cf. Romans 13.1: non est enim potestas nisi a Deo ‘there is no power but of God’.) Does it depend from god directly (depend[e] t [a deo] immediate, i.e. rather than by the intermediary of the pope)? Where is it situated at present (ubi hodie exsist[i]t)? Nicholas notes that those are questions that many contemporary scholars address differently and at length (istae quaestiones varie ac prolixe per multos modernos doctores habentur). He declares that he will not address them himself, ¹²⁷ but then does so anyway—he was evidently writing in a hurry, which may explain certain inconsistencies. ¹²⁸ He begins by debunking the Donation of Constantine (the eighth‐century fake according to which the emperor Constantine I bestowed the imperial prerogatives in the west on the pope when he moved the capital of the empire to the Bósporos) and the notion of the translatio imperii or transferral of the empire (according to which when pope Leo III crowned the Frankish king emperor in 800 he was taking away the imperial dignity from Constantinople, a notion developed from the eleventh century onwards). Nicholas rejects both on the grounds that his extensive research failed to yield any mention of either in ancient sources where one would expect references to them (more famously, but a little later, the humanist scholar Lorenzo Valla demonstrated the inauthenticity of the Donation of Constantine on both historical and philological grounds).

(p. 408 ) This was tantamount to positing the independence of the empire from the papacy. ¹²⁹ But what was the emperor needed for? According to Nicholas, the empire as such was not universal, since there were kings and princes not subject to the emperor. Nonetheless, the emperor was the guardian of the universal faith and the keeper of the universal statutes (universalis fidei custos ac statutorum universalium conservator: a position close to that outlined by patriarch Antony in his letter to Moscow). In that capacity he could, according to Nicholas, even pass laws (leges) that bound those who otherwise ‘de facto or by privilege’ did not recognize the suzerainty of the empire and who therefore ‘perhaps’ were not bound by other imperial laws. ¹³⁰

As was warranted by the circumstances in which this treatise was written or at least finished, Nicholas discusses at some length the role of the emperor at the council. He recalls that in previous ages this role was very important—his exemplar here is, in particular, the Eighth (and last) Ecumenical Council, which met in Constantinople in 869/70 and which was presided over by the emperor Basil I; Nicholas quotes extensively from its proceedings. Of course Sigismund, too, was conspicuous enough at both the council of Konstanz and the council of Basel. Nicholas nevertheless concedes that the emperor no longer had the power (potestas, i.e. the capability combined with the right) to ‘bind and restrain all’ (cum imperatoris potestas hodie non omnes sicut quondam stringat et compescat). Consensus should therefore be achieved through consultation with ‘the other princes and the kings that are in christendom’ (per principum etiam aliorum et regum, qui in Christianitate sunt, advocationem). ¹³¹

Despite the responsibility of the emperor for christendom as a whole, for Nicholas the empire as such was clearly, in practice, coterminous with Germany. He is not pleased with the political situation there, which he discusses with noticeably less academic detachment than he does christendom as a whole.

There no longer is any concern for the commonwealth [or: the public good] …the laws are all woven of spider webs, hardly capable of catching even the smallest locusts. …All control has ended. Rebels go unpunished. And as the empire has diminished, many powerful princes have sprung up by usurpation. …Oh the supreme blindness! The princes should not believe that, having enriched themselves out of the property of the empire, they will stay rich for long. For if each is only concerned to increase his property while the empire is in the process of vanishing, how can this not end in the destruction of all? Absent a greater power [potentia] that preserves and pacifies the empire, jealousy and this same ever‐growing greed will cause wars, schisms, and divisions …the hierarchical order will lapse. There is no first man for all to turn to. And where there is no order there is chaos. And where there is chaos no one is safe. And thus while the nobles quarrel among themselves, leaders of the people rise up who seek the law only in their own weapons. Just as the princes devour the empire, so the leaders of the people devour the princes. …And people will search for the empire in Germany, and they will not find it there. And as a result others will seize our lands, and we shall be divided among ourselves, and so shall become subject to some foreign nation. ¹³²

(p. 409 ) It is a little difficult to see what, concretely, Nicholas is talking about in this famous passage. There certainly was no strong central power in Germany, but there never had been, and even though among the German princes no doubt there were few paragons of patriotic self‐abnegation, nevertheless the situation of the empire in the 1430s could hardly be described as lawless chaos. Feuds might occur, but there was no generalized conflict among the princes, nor any generalized popular rebellion against them. Nicholas probably has in mind the Hussites in Bohemia, much discussed at Basel. Defying the established social order, they had indeed set up a radical, broad‐based regime that proved militarily invincible, and capable of raiding into the empire far beyond Bohemia itself. That Nicholas was thinking of them is also suggested by his use of the term schismata 'schisms’, which implies a religious dispute: the Hussites rejected the universal church on theological grounds. But this movement remained confined to Bohemia; elsewhere there was neither any conspicuous contestation of the established social order nor, as yet, of the universal church. The quoted passage oscillates between the perfect tense, the present tense, and the future tense. It is evidently only partly inspired by the actual political situation: much of it is an apocalyptic extrapolation.

Nicholas’ negative appraisal of the empire stands in a tradition. We saw that both Adso of Montier‐en‐Der in the tenth century and Engelbert of Admont in the early fourteenth century regarded the empire as largely destroyed. Many more adherents of that view could be adduced. In the 1140s, Otto of Freising, an uncle of the emperor Frederic I, remarked that the Roman empire, once supreme, had of late become almost the least respected realm of all; ¹³³ and Jordanus of Osnabrück, very much like Nicholas, already in the third quarter of the thirteenth century warned ‘the German prelates and princes [presules et principes Germani]’ to be careful ‘lest in their ambition for temporal power they claim for themselves and usurp rights and property of the empire’, hastening its downfall and thereby the coming of the antichrist. ¹³⁴ Like Nicholas (o caecitas maxima! ‘oh the supreme blindness!’), Dante, in castigating what he saw as the self‐seeking particularism of those of his fellow Italians who opposed the emperor‐designate Henry VII, called them ‘blind’ (Nec advertitis dominantem cupidinem, quia ceci estis; La cieca cupidigia che v'ammalia). Henry himself took a similar view: proterviunt inconsulte , ‘thoughtlessy they mock us’. ¹³⁵

Nor, as we likewise saw, was this type of complaint confined to the empire. Writing roughly at the same time as Dante and Engelbert, Pierre Dubois thought that ‘the vice of disrespect for the welfare and interest of the commonwealth has up to now been more (p. 410 ) strongly developed in the kingdom of France than in other parts of the world’; and in the late fifteenth century, Philippe de Commynes lamented that ‘our realm [France] …is more oppressed and persecuted by this situation [of recurrent violent conflict among the major princes] than any other lordship that I know.’ ¹³⁶ Pierre and Philippe were active in the reign of what are now normally considered 'strong’ kings, Philip IV and Louis XI. The common denominator of all those utterances is that they express a longing for a strong monarchy. This must have helped efforts by the crown to enhance its position, while at the same time it suggests that few observers in pre‐Reformation christendom were impressed by the power of the crown (any crown) as it then was. It seems that ‘real’ monarchy was widely considered a goal whose (future) attainment would finally cure society of many much‐resented ills.

There is a surprising resemblance between Nicholas’ critique of the empire and Thomas Hobbes’ famous theory about the need for subjection to a 'sovereign’ as the only remedy for the war of all against all that, for Hobbes, in the absence of such a sovereign is inescapable. Nicholas insists on the need for a ‘greater power’ (potentia maior: power in the sense of physical power) to keep everybody else in check, a ‘first man’ (primus) to whom recourse can be had, failing which there will be chaos (confusio) and wars (guerrae) on account of the selfishness (cupiditas) of individuals all intent on expanding their possessions (curantibus omnibus sua augmentare), leading to a situation where no one was safe (nullus tutus). As Hobbes noted but Nicholas did not, this kind of position was incompatible with Aristotle's axiom that man was by nature a social being. ¹³⁷ Thomas Aquinas and Giles of Rome also called for a strong ruler capable of coercing his subjects, but, well‐versed in, and respectful of, Aristotle's political thinking, did not go so far as explicitly to make coercion the necessary precondition of an orderly social life. Nicholas, by contrast, despite great erudition displayed throughout the De concordantia seems only to have ‘discovered’ Aristotle when he wrote the introduction to Book III of the treatise, apparently the last part to be finished; and his acquaintance with Aristotle seems to have been limited to such quotations from the political writings of the philosophus as he found in Marsilius of Padua. Elsewhere his political thinking is far removed from Aristotle's. Thus, according to him

The force of the law lies in coercion. Coercion is preserved by [physical] power [potentia]. If this is taken away, control by the law and therefore peace and justice will not last long, since we persevere in what is forbidden and the imagination of our heart is evil from our youth [an allusion to Genesis 8.21]. ¹³⁸

This bleak view of human nature is again paralleled in Hobbes, as is the notion that no law can exist unless backed by coercion. Aristotle, as we saw, held the opposite view, insisting that ‘the law has no power to secure compliance other than habit.’ ¹³⁹

(p. 411 ) Concretely, Nicholas pleads for the ‘restoration’ of an imperial army, mistakenly assuming that the tenth‐century (east Frankish) empire, which he regarded as a model, had already possessed it. This standing army was to be financed from taxes and to have the task not just of defending but also of policing the country, in particular to combat robbers; that, Nicholas thought, would be cheaper and more efficient than the current situation, where this task fell on local power‐holders. ¹⁴⁰ Nicholas also condemns feuds as incompatible with Roman law; he wants to see them abolished in favour of compulsory jurisdiction. ¹⁴¹ That he devotes quite a lot of space to such proposals may be due to the fact that the emperor Sigismund had called the Reichstag (imperial diet) to Basel for January 1434, to sit in parallel with the council; the diet was in particular to discuss the problem of ‘peace and justice’, the suppression of violence and crime. Nicholas’ proposals clearly go in the direction of a central power with a monopoly on legitimate violence. Nicholas wants that power controlled by an assembly representative of the country—such as, precisely, the 1434 Reichstag. He is, however, vague on the composition and prerogatives of that controlling assembly. ¹⁴²

In contrast with Engelbert and Dante, who ascribed to the emperor only those responsibilities that the lower‐ranking power‐holders or communities could not exercise themselves, for Nicholas the power of the emperor is not residual but absolute, or in any case should be. Nicholas thought that when, in 962, Otto I received the imperial crown in Rome—thus again re‐establishing the western imperial dignity after the Carolingian effort to do so had failed—the Romans (in the sense of inhabitants of the city) accepted that he should have vera suprema imperialis potestas absque diminutione et conditione, truly supreme imperial power without limitation or condition. ¹⁴³ This description of imperial power sounds very much like what Jean Bodin would later call souveraineté. By contrast, the notion, which Engelbert or Dante took for granted, that there should be power‐holders and communities below the ‘first man’ and wielding power in their own right for Nicholas was a source of anarchy.

So we have Nicholas (and Hobbes) claiming that society and law cannot subsist in the absence of coercion and Aristotle claiming the opposite; and we have early fourteenth‐century authors accepting unquestioningly the existence of intermediate power‐holders whose power was not delegated by the central power when for Nicholas (and Hobbes would have agreed) this is a recipe for chaos and a threat to the social order. We are clearly dealing here with political preferences, or fashions, that were not dictated by any evolution of society itself. It is not as if what worked (more or less) for the pólis—law enforced almost exclusively by peer pressure and social control rather than coercion—no longer worked in fifteenth‐century Germany (or seventeenth‐century England), or as if the fourteenth century, with its concept of society as a hierarchy of autonomous communities topped by a concomitantly weak central power, was characterized by greater social instability than the fifteenth century, when this concept began to be rejected and 'sovereignty’ (to call it by its later name) of the ruler came to be seen as desirable. Of course, calling for strong, coercive governance came more easily as few had actually experienced any such thing.

(p. 412 ) The autonomous communities envisaged by the early fourteenth‐century scholastics could be treated simply as given, since their genesis could be assumed to be spontaneous. The kind of central authority envisaged by Nicholas needed to be set up. On this model, political communities came into being by virtue of a treaty between the ruler and the subjects. Remarkably, Nicholas depicts the ruler as both a larger‐than‐life ‘father’ and a ‘creature’ of the ruled, in a manner reminiscent of the illustration on the famous title page of the first edition of Hobbes’ Leviathan of 1651—a giant composed of the bodies of many individuals and representing Hobbes’ 'sovereign’. The relevant passage, which also recalls Hobbes in other respects, would have benefited from revision by Nicholas, for which he apparently found no time. It is so complex that Nicholas himself lost his way in it grammatically, a reflection, perhaps, of his own difficulty with the subject matter also.

Now if we recall that which we discussed earlier, that every superior power [superioritas] properly ordained arises from the elective unity of voluntary subjection, and that the people [populus] partakes of that divine seed by virtue of the equal birth and the equal natural rights shared by all men, so that every [legal] power [potestas], which, like man himself, takes its origin from god, is considered divine if through their shared consent it arises from the subjects, so that the person established in this fashion, and in exercising his princely office as it were carrying in himself the will of all, should be called a public and shared person and the father of the individuals, governing everything with rightful, rule‐bound, properly ordained might [potentia] and without pomp or arrogance, and knowing himself to be the collective creature of all his subjects, he is the father of the individuals. ¹⁴⁴

Despite the convoluted phrasing, the basic theses are clear: the subjects, whose natural equality is emphasized, create the ruler by means of their voluntary subjection to him. So on the one hand that ruler is responsible to them, but on the other all power or might (potentia) is his, governing ‘everything’ (cuncta gubernans); and the whole arrangement is regarded as enjoying ‘divine’ legitimation. The ruler exercises his function less as a human being than as an abstract ‘public’ person, rather like Hobbes’ ‘Artificial Man’, and in line with Hobbes’ assertion that ‘every Subject is Author of every act the Soveraign doth.’ ¹⁴⁵

4.2.5.2 Enea Silvio Piccolomini, Frederic III of Habsburg, and Maximilian of Habsburg

When Nicholas presented his treatise to the Basel council, the participants included one Enea Silvio Piccolomini (1405–64). Elected to the papacy in 1458, he was later to (p. 413 ) make Nicholas a member of the curia and his deputy when he was not in Rome. ¹⁴⁶ He evidently read De concordantia, since bits and pieces of that treatise reappear (without acknowledgement) in his own treatise De ortu et auctoritate imperii Romani ‘On the origin and authority of the Roman empire’. This treatise has the form of a long letter, dated Vienna, 1 March 1446, to the then king of Romans, Frederic III, a Habsburg whose secretary Enea Silvio was at the time. ¹⁴⁷ Further borrowings, likewise unacknowledged, come from Engelbert and Dante. In the latter instance it is, however, possible that Enea Silvio lifted the relevant material not direct from the Monarchia of Dante, but from the eponymous treatise of his own teacher, the legal scholar Antonio Roselli, who had himself appropriated much of the work of his predecessor without ever mentioning his name. ¹⁴⁸

Enea Silvio evidently drew the inspiration for his title from Engelbert, whom indeed he partly follows on the ‘origin’ (ortus) of the early kingdoms in general and then of the monarchia in the sense of lordship over the whole world; however, there are also telling departures. According to Enea Silvio, after their expulsion from the Garden of Eden men at first lived as scattered individuals, but then discovered the advantages of communal life. To resolve disputes they found it necessary to subject themselves to a common arbitrator, who later received the title king. Led by kings, the original ‘nations and peoples’ (gentes nationesque) were not interested in expanding their territory beyond their own homeland (intra suam cuique patriam regna finiebantur). ¹⁴⁹ But soon quarrels erupted between the kings over questions of territory and jurisdiction. Since they did not recognize each other's authority, those quarrels could only be settled by the sword. This led to the creation of ‘what the Greeks call monarchia, and we ourselves [nostri] call imperium’—for this was the only way to ensure a general peace (Nec enim aliter pax universalis haberi poterat). ¹⁵⁰

First there was the empire (imperium) of the Assyrians under Nínos, later ruled by the Medes and Persians, the empire (Enea Silvio refers to it as a principatus) of the Greeks under Alexander, and the empire (dominium) of the Carthaginians. Each is introduced with hinc ‘thence, therefore’, suggesting that each is seen as a separate instantiation of the basic phenomenon rather than as part of a pre‐ordained succession as in the traditional interpretation of the prophesy of Daniel—which in any case knew of no empire of the Carthaginians. That Enea Silvio wanted to stay clear of Daniel here is also indicated by the fact that he does not employ the word imperium throughout but varies his terminology. Unlike Engelbert, Enea Silvio does not regard the phenomenon of empire as, at least in origin, the product of (reprehensible) passion rather than reason, instead presenting it as divinely willed and as a product of reason because empire was the precondition of general peace.

Whereas Engelbert and Dante felt compelled to prove the legitimacy of Roman rule against the accusation that the Roman conquests were unjust, Enea Silvio—and, for (p. 414 ) that matter, Nicholas—completely ignored that whole theme. Ubiquitous in the early fourteenth century, in their day this particular debate had clearly gone out of fashion. Nor did Enea Silvio have any use for Engelbert's eschatologically motivated fatalism, based in part on the prophesy of Daniel and according to which the decline of the empire would ineluctably continue until the coming of the antichrist. Engelbert's treatise was about ‘the origin and [future] end of the Roman empire’: even though the translation ‘the origin and purpose of the Roman empire’ is likewise possible, the former sense is unavoidably present, too. By contrast, Enea Silvio, striking a less ambiguous and more positive note, called his own treatise ‘On the origin and authority of the Roman empire’.

Since the Assyrians, Medes, Persians, etc. ‘never brought the entire world under their control [nunquam sibi totum orbem subjecissent] and therefore were unable to establish a general peace, it pleased nature, who nourishes the human race, or god as lord and leader [rector] of nature, to bring the Roman empire into being’. ¹⁵¹ The legitimacy of the Roman empire is derived by Enea Silvio from the usual staple of arguments that we encountered before. Thus, there is the indispensable observation that Christ was born in the reign of Augustus, as well as a reference to the reply given by Jesus to Pontius Pilate ‘Thou couldest have no power at all against me, except it were given thee from above’ (John 19.11). This, as we saw, is also adduced by Dante (Monarchia 2.11), though Enea Silvio may likewise have found it in Jordanus of Osnabrück. Acquaintance with that latter author is suggested by the 'spin’ that Enea Silvio puts on the notion that the Roman empire would not end until the coming of the antichrist—interpreting it, like Jordanus, as a special honour bestowed by god on the Roman empire. ¹⁵²

What most sharply distinguishes Enea Silvio from both Engelbert and Dante is his emphasis on the fullness of the imperial power, a subject on which he seems to be following in the footsteps of Antonio Roselli. As in the case of the original realms discussed at the beginning of the treatise, so, too, the universal monarchia should be ruled by a single person, which according to Enea Silvio is a dictate of ‘natural reason’. But who could this single person be other than the king of Romans, long known to have been in possession of that office (quem diu constat in possessione ejus fuisse)—presumably an allusion to the Roman law concept of usucapio, according to which a possession must be considered lawful if sufficient time passes without its being contested effectively. ¹⁵³ Enea Silvio concedes that there were people unwilling to submit to the empire, yet points out that even so no one had yet thought of calling someone other than the king of Romans lord of the world (mundi dominus). This incidentally establishes an implicit equivalence between the ‘king of Romans’ (as the addressee of the treatise, Frederic III, as yet was when it was written) and the ‘emperor of the Romans’ (as Frederic became by virtue of his imperial coronation in 1452). Enea Silvio also follows the example of Engelbert in simply ignoring the question of whether the pope had some kind of suzerainty over the emperor, thus implicitly denying that he did. (Nicholas, for his part, expressly states that the ruler of the empire had the same rights whether he had received the imperial coronation or (p. 415 ) not, and thought that the point of having the pope bestow the imperial title proper was merely to encourage the incumbent to seek a close relationship with the Holy See.) ¹⁵⁴

In contrast with Engelbert, but in line with the trend to do away with traditional limitations on the power of the supreme ruler, Enea Silvio rejects the notion that anyone could be exempted from the suzerainty of the empire by some privilege. This is precisely because the ruler of the empire is absolute, all‐powerful: according to Enea Silvio, no more than god himself could the emperor create someone who was his equal. Moreover, this would be in contradiction with the natural purpose of the empire, to preserve peace (pacem tenere) and to apply justice (justitiam distribuere). A plurality of supreme powers (multitudo summarum potestatum) could not do that. ¹⁵⁵ The existence of other princes is a given even for Enea Silvio, but he insists that by rights the emperor exercises a suzerainty over them that is not merely nominal.

We do not …contest that kings and other princes have great power [potestas], but we say that they are under the empire [eos esse sub imperium dicimus], and that it is incumbent on the Roman prince to take them to task if they become tyrants. We affirm that all quarrels of the kings must be submitted to the emperor [cesar] and that for the sake of the common welfare [salus communis] all are obliged to defer to the emperor [imperator], to assist him militarily when called on to do so, to contribute expenses, to send troops, to permit their passage, to furnish supplies, and not to evade anything that the imperial majesty [cesarea majestas] has enjoined, granting to their lord the emperor [imperator] the same obedience that they demand of their own subjects. ¹⁵⁶

For Enea Silvio, there was no need for obedience to be counterbalanced by representation as in Nicholas. Enea Silvio moreover rejects the principle of the ‘feudal’ order empowering subjects or vassals to resist their lord if in their view he acted unjustly. Such unjust behaviour by the lord was the will of god and often a punishment for the sins of the subjects. The emperor, and he alone, could make laws binding on everyone, and he alone could interpret or repeal them. He himself was not bound by laws, legibus solutus. He should normally respect them, but could deviate from them if equity suggested it. ¹⁵⁷ For Enea Silvio, there could be no appeal against a decision of the emperor, just as according to a legal principle of the church (challenged, it is true, by the conciliar movement) the pope could be judged by no one. ¹⁵⁸

Like Nicholas, Enea Silvio sees the imperial office as the secular equivalent of the papal office: ‘just as in spiritual matters the individual …bishops are subject to the Roman pontiff, so, too, is it clear that secular persons of whatever kind are subject to the Roman prince [sicut in spiritualibus Romano pontifici singuli …prelati subjecti sunt …sic et Romano principi temporales quoslibet liquet esse subjectos]’. ¹⁵⁹ (Curiously this postulates authority of the pope only over the leaders of the church hierarchy, whereas the emperor is given authority over all laymen.) As Antonio Roselli ¹⁶⁰ and Nicholas ¹⁶¹ (p. 416 ) had done before him, Enea Silvio applies the doctrine of the church as the mystic body of Christ (corpus mysticum) to the ‘commonwealth’ or res publica: this, too, is treated as a mystic body (or perhaps rather: the same mystic body as the church), with the emperor as its head (princeps, qui caput est mystici rei publice corporis). ¹⁶² As was the case with early‐fourteenth‐century writers like Engelbert or Pierre Dubois, the ‘commonwealth’ tout court is still christendom, clearly still the natural interpretation since Enea Silvio saw no need to define what he meant.

Antonio Roselli had adapted arguments developed by Dante to assert the primacy of the emperor to defend the primacy of the pope against the challenge of conciliarism. Enea Silvio re‐adapts those arguments to assert the primacy of the emperor once more. ¹⁶³ This is another instance of how concepts of rule that in the post‐Reformation period became ingredients of the general theory of princely rule now commonly called ‘absolutism’ were shaped in the pre‐Reformation period in the debate on the respective rights, and mutual relationship, of the emperor and the pope, as well as on their role with regard to christendom generally. With regard to christendom as a whole, the emperor as presented in the political thinking of Nicholas and Enea Silvio remained a theoretical construct. Yet the two authors are instructive in that they illustrate how ideas of rulership and political organization had changed since the time of Engelbert and Dante (especially as Enea Silvio at least was actually conversant with the thinking of those two). The ruler as conceived by Nicholas and Enea Silvio is the source of all power rather than merely wielding what power could not be wielded at a lower level of political organization, more particularly the lower communitates (household, village, city, kingdom) of Engelbert and Dante. Thomas Aquinas and Giles of Rome (cited approvingly by Nicholas) ¹⁶⁴ had already stressed the power and importance of the king, but while they completely ignore any lesser lords than the king, in their thinking the self‐organizing communitates are likewise prominent. By contrast, Nicholas and Enea Silvio do away with the communitates, which simply drop out. The supreme ruler alone now runs the show, though (in Nicholas) he faces some kind of representative assembly of the subjects. Of course, in the nature of a theory of empire rather than of kingship neither Nicholas nor Enea Silvio could ignore that the king of Romans was not the only king in christendom. But Nicholas, having established a rather vague superordination of the emperor over every other ruler in christendom, whether or not they were subject to the empire, in matters regarding the common faith, then concentrates on German politics, which he had clearly more at heart; Enea Silvio prefers the intrinsic logic of his theory over contemporary political reality in simply denying that anybody at all could be not subject to the empire. In a similar spirit, as pope he issued a stinging condemnation of conciliarism (papal bull Execrabilis, 1460).

John Ruggie has observed that the ‘[t]he concept of sovereignty …was …the doctrinal counterpart of the application of single‐point perspectival forms [in the figurative arts] to the spatial organization of politics’ ¹⁶⁵ —precisely around the time when Enea Silvio composed his treatise. His political thinking has the clarity and systematic consistency of the depictions of ideal, classical‐inspired architecture popular among Renaissance artists. As tended to be the case with those ideal designs, too, it found at best an imperfect echo (p. 417 ) in projects carried out in the ‘real world’. On the other hand, of course there is a marked resemblance between architectural projects of that period that were actually carried out and those ideal designs—they share the same, novel style, superseding the Gothic style, which by the sixteenth century gave way to it everywhere. Interestingly, late in his life Enea Silvio himself ventured into town planning: having become pope he renamed his Tuscan birthplace, the village of Corsignano, Pienza (i.e. Pius‐Town, after the name he took as pope, Pius II), elevated it to an episcopal see, and started rebuilding it as an ideal Renaissance town, with various palaces clustered around, and oriented towards, the new cathedral at the centre.

Frederic III never became the all‐powerful emperor envisaged by Enea Silvio. On the throne for over half a century, for much of his reign he was contested from within his own dynasty and had a hard time even hanging on to his hereditary lands, to say nothing of exercising a powerful influence as emperor. The great question of his reign was the future of the Burgundian dominions. We saw that king Charles V of France managed to secure the hand of the daughter and only heir of the last count of Flanders for his brother, whom he created duke of Burgundy. In this way, two of the great principalities forming part of the French kingdom, Flanders and Burgundy, were united under a member of the French royal house—a solution preferable by far, for king Charles, to the prospect of a marriage between the Burgundian heiress and a son of Edward III of England, which would have resulted in two of the great French principalities, Flanders and Aquitaine, being held by the English royal house. But the Burgundian branch of the Valois dynasty thus created subsequently embarked on a dynamic career of its own, collecting more and more dominions stretching from Holland down to Burgundy proper, even though they were not all contiguous. (This accretion of dominions occurred mainly through inheritance and purchase. Moreover, Burgundian support of the English claim to the French crown was paid for with additional fiefs granted by the English crown, possession of which was then confirmed in return for switching support back to the French king.) Sandwiched between France and Germany, all those dominions were either French or German fiefs, the duke of Burgundy theoretically remaining a vassal for all the lands that he ruled and lacking any dominions held in his own right—but he behaved like a king, his court easily rivalling any royal court of the period in terms of wealth and refinement.

However, like the count of Flanders a century earlier, the last duke, Charles ‘the Bold’ (1433–77, in power since 1454), only had a female heir, Mary, ardently sought in marriage by both the French crown and the house of Habsburg as a way to recover the fiefs belonging to their respective kingdoms—while, if possible, of course hanging on to the rest as well. When Charles was killed in the battle of Nancy in 1477, Louis XI of France hastened to occupy those Burgundian fiefs that the duke had held from the French crown. Shocked at this, the Burgundian court at last opted for Frederic, whose son Maximilian got to marry the orphaned Burgundian heiress. Their son, heir to the combined Habsburg and Burgundian hereditary lands despite the loss of some Burgundian fiefs to the French crown, would later marry the heiress of the Spanish realms (and the Spanish overseas colonies), providing their grandson, the emperor Charles V, with dominions ranging from the Philippines to the Americas, and in which famously the sun never set. It should be noted that—as had also been the case, on a smaller scale, with the Burgundian lands themselves—this spectacular expansion of the Habsburg dominions was not achieved by conquest. Though unusual in its dimensions, the phenomenon as such is entirely typical (p. 418 ) of the ancien régime, where acquisitions by inheritance tended throughout to be vastly more extensive than acquisitions by conquest.

Being nominal lord of a vast array of dominions did not mean that the ruler was necessarily very powerful in those lands; moreover, his position would be different in each of his separate dominions. Yet in a highly status‐conscious culture, possession of many dominions was prestigious, a source of pride and envy. More than his father Frederic, whom he succeeded as king of Romans in 1493, Maximilian was not only imbued with the magnificence of his position, but, in the manner that became typical of princes in the post‐Reformation ancien régime, was keen on outward expression of this magnificence.

In 1508, Maximilian commissioned his funeral monument, employing the best available artists. Perhaps the largest and most magnificent monument of any prince in the history of christendom, at his death in 1519 it was still unfinished; moreover, it turned out that even so the floor of the church of Saint George at Wiener Neustadt, where Maximilian is buried, would not support the weight of its many bronze statues because of the crypt underneath. The monument was to have comprised a funeral procession of forty more than life‐sized ancestors and relatives of the deceased, of which only twenty‐eight were carried out; as were thirty‐four busts of pre‐ and early christian Roman rulers and twenty‐three statues of saints with whom the house of Habsburg was considered to have a special affinity. (In the 1570s, the emperor Ferdinand I at last had the unfinished monument installed in the newly‐built Hofkirche or palace church at Innsbruck.) Only a decade and a half separate the planning of this gigantic Renaissance monument, which has nothing Gothic at all, from the magnificently carved, but perfectly conventional and determinedly Gothic marble tomb of Frederic III in Vienna cathedral—which in comparison looks humble indeed; but Frederic, aged 77 at his death in 1493, was the product of an earlier age.

In Germany, Maximilian, because of his love of chivalry, is popularly known as ‘the last knight’. But this is a romantic misconception (chivalrous ritual, such as tournaments, remained popular throughout the sixteenth century). Rather than being a remnant of a dying age, he really epitomizes the prince of the future, much more imbued with his own importance than was typical of pre‐Reformation Latin christendom. This change took place despite the fact that Maximilian was hardly a more powerful figure, within the empire or within Europe, than his predecessors. Within the empire, he shared power with the estates of the empire, that is the greater princes and the free cities. Within christendom at large, he was the senior of a dynasty whose extensive possessions spread over different parts of Europe and which had family ties with practically everybody who was anybody in christendom—thus, as just one aspect of this, daughters or granddaughters of Maximilian himself were, at one time or another, married to the king of France, the king of Portugal, the next king of Portugal, the king of Denmark, Norway, and Sweden, the king of Bohemia and Hungary, and the duke of Savoy. But while this made Maximilian a key figure with a great deal of influence, it did not mean that he could lay down the law to anybody—not within the dynasty, nor within the empire or even simply the Habsburg hereditary lands, nor within christendom at large.

It was in the same year 1508 that Maximilian unilaterally assumed the imperial title. In a proclamation informing the estates of the Holy Roman Empire of this step, he explains, with apparent humility, that he was unable to obtain the imperial coronation (p. 419 ) from the pope, ‘owing to our small power and great opposition, the like of which no Roman king has ever encountered’. Appealing to German national sentiment, he protests that the behaviour of the Holy See was unfair and that, without prejudice to any papal prerogatives and in the hope that the coronation would yet take place, he felt compelled to adopt the imperial title ‘for the sake not just of our own honour but rather for the sake of confirming and preserving the Roman imperial dignity, in honour of us all and of the German nation [nicht allein um Unserer Ehre willen, sondern mehr zur Bestätigung und Erhaltung des Römischen Kayserthums, Uns allen und Teutscher Nation zu Ehren]’. ¹⁶⁶

Less than a decade before Martin Luther set in motion the movement that was to end the unity of Latin christendom and the authority of the pope over all of it, this document already makes clear that a new age had begun. Two hundred years earlier, in the reign of Henry VII, assuming the imperial title without the requisite ecclesiastical ceremony in Rome was utterly unthinkable. That title in itself was considered of such moment that his adversaries tried to block the ceremony by force of arms, occupying Saint Peter's cathedral. Having obtained the coronation anyway, Henry in his circular informed all christendom of that fact, employing, of course, the common language of christendom, Latin, and at length invoking universal christian politico‐religious doctrine. In 1508, we find his successor Maximilian deciding that if he cannot have the ceremony, that, with due respect to the Holy See, was no reason not to adopt the imperial title—for which his position as king of Romans made him the automatic and sole candidate. No doubt the courts of christendom were also informed of this step. But the place of the 1312 circular is here taken by a communication internal to the Holy Roman Empire, employing the German language, and appealing to German patriotic pride. In striking contrast to the 1312 circular, in the 1508 proclamation christian doctrine plays no role at all.

This is not to say that the traditional ideology of the empire with its universalist and strongly religious orientation was abandoned by Maximilian—it is still found in his reign, and under his successors well into the seventeenth century. But the emphasis on this kind of discourse decreased steadily. Conversely, the overt invocation of German national sentiment not only marks another break with the past—if I am not mistaken, Maximilian was the first emperor to resort to it—but it caught on massively; one is tempted to say, with a vengeance. By the mid‐sixteenth century, the political discourse of the Holy Roman Empire, including, but far from limited to, official utterances by the crown, ¹⁶⁷ became replete with references to the ‘beloved fatherland of German nation’ (equated with the empire), with which it remained saturated throughout the ancien régime. One more emperor, Charles V, was crowned by the pope, in 1530; after that, all took the imperial title from the moment of their accession as sole rulers (having mostly been elected and crowned as king of Romans in the lifetime of their predecessors).

Beginning in the fifteenth century, throughout Europe the role, or at least the profile, of princes increased. It really does appear that, as John Ruggie has surmised, the major, fundamental change went on in people's heads—Ruggie speaks of a ‘broad transformation in the prevailing social episteme’—and that it occurred quite fast, within a generation or two. ¹⁶⁹

According to Ruggie, the main agent of change was 'social empowerment’, a new conception of politics creating acceptance for a centralization of power in the hands of the prince in much the same way as painters adopted single‐point perspective—in response neither to any functional need, nor to coercion, nor to any ‘technology’ not available earlier. Nevertheless, certain trends in the economic and military evolution of European civilization no doubt promoted the increasing focus on centralized princely governance and are worth recalling, briefly and without any claim to being exhaustive. Cause and effect are often hard to distinguish here. In most instances, what we have is probably best described as a cumulative, self‐reinforcing process, where the ‘transformation of the prevailing social episteme’, of the prevailing way of looking at the social world, encouraged certain types of material change that in its turn consolidated a new climate of thought. The new thinking itself, on the other hand, remained permeated with ideas on the role of power‐holders imprinted on western civilization by christianity.

4.3.2.1 Economic and military factors

In the economic sphere, the overall expansion characterizing the period after the turn of the millennium continued and probably intensified in the sixteenth, seventeenth, and eighteenth centuries, helped by the fact that Europe increasingly dominated the rest of the world: new crops, raw materials, trade goods, and markets kept stimulating trade and thus monetization. In the conditions described earlier, more trade also meant more credit, an expansion of the financial system of Europe accompanied by its increasing (p. 423 ) sophistication. This general expansion meant that more and more wealth was up for distribution within Europe.

In the latter part of the ancien régime, many of the proud self‐governing towns of Latin christendom went into decline, not least economically. One reason for this was the tendency to move artisanal production to the countryside, where it was not controlled by guilds: this meant that there were no restrictions on the nature or quantity of what could be produced, or by whom. Since workers did not have to be approved by the guilds, and women and children were not barred, either, the workforce was greater and the cost of labour, less. Entrepreneurs mostly still resident in the towns supplied the raw materials and collected and marketed the output, with production itself put out to scattered rural households. ¹⁷⁰ This was true for example of the textile industry—woollen and linen cloth—of northern France in the seventeenth and eighteenth centuries. Wealthy citizens of towns like Beauvais, such as Gabriel Motte, whom we encountered earlier, distributed raw materials partly imported from far away (such as fine Spanish wool) to villagers and then sold the finished cloth throughout Europe and even in the Americas. ¹⁷¹ On the plateau separating the towns of Amiens, Aumale, and Beauvais, many more looms could be found than ploughs. ¹⁷² This situation—full‐time farmers being outnumbered in the countryside by artisanal workers—was likewise found throughout much of the rest of late ancien régime Europe.

Even the greatest cities of the pre‐Reformation period had usually been dominated by merchants. After the Reformation, such cities were increasingly rivalled by princely residences. Some princely residences, like Paris, London, or Naples, had been great cities before, but they were now joined by many that before the Reformation had been of little significance. Thus, when, in the fifteenth century, the Hohenzollern family was enfeoffed with the electorate of Brandenburg, all of whose towns were quite modest, they chose to establish their main residence at Berlin—then a settlement of perhaps some 6,000 people, which actually put it among the large towns of the electorate. By 1730, the population of Berlin had risen to about 72,000, approaching 100,000 in 1760. ¹⁷³ This was a lot less than Paris, London, or Vienna had at that time. But since nothing in its geographical position in the middle of a thinly settled land of poor soils and few other natural resources predestined Berlin ever to reach this size, it exemplifies the effect of serving as a princely residence particularly well.

We have already discussed the crucial role played by monetization as an enabling factor for centralizing power. As a result, from the late pre‐Reformation period onwards political power was increasingly paralleled by financial power. Armies made up of vassals gave way to armies made up of mercenaries, and successful warfare increasingly demanded manpower, weaponry, and infrastructure (e.g. fortresses) on a scale putting it beyond the means of minor lords. The evolution of fortifications well exemplifies the phenomenon. The vertical walls of pre‐Reformation castles and towns were easy to breach with heavy artillery. To make them reasonably difficult to overcome, a new type of fortifications was necessary, the so‐called trace italienne first encountered in fifteenth‐century Italy: relatively low, sloping walls reinforced with earthworks, broken up by complicated, (p. 424 ) many‐angled bastions; a wide glacis, secured by a wide ditch or moat and ancillary fortifications beyond it, served to prevent attackers from getting close. Fortifications of this type were difficult, or at least uneconomical, to execute on a small scale and typically enclosed whole towns; where they were built on a green field site, such as some of the fortresses constructed by Vauban to secure the frontiers of France under Louis XIV, this often actually amounted to the foundation of new towns. So huge were fortifications of this new type that they often took up more space than they protected: for example, the relation in terms of area between the fortifications of eighteenth‐century Mainz and the settlement at their centre was 4:3. It has been estimated that the cost of fortifications rose by a factor of thirty in the course of the seventeenth century. ¹⁷⁴

Armies (and to a lesser extent, navies) expanded, too.

The sixteenth century, at least from the 1530s onwards, had seen an unprecedented growth in the size of armies; and this was to continue throughout the century which followed. The Spanish army, perhaps a mere 20,000 in the 1470s when Spain's rise to the military leadership of Europe was about to begin, had reached 300,000 in the late 1630s …France, which was maintaining an army of about 50,000 in the 1550s, had one three times as large in the 1630s. ¹⁷⁵

Building and maintaining armies, navies, and fortresses was expensive in itself, and using them was more expensive still. Again the new fortifications were most responsible for this.

A long siege ate up money, materials and often men in a way that even the greatest battle hardly ever did. …A great siege was, with the possible exception of the building of a great canal, the biggest engineering operation known to the age [the 17th and 18th centuries]. …[It] could therefore strain the resources of the greatest states so seriously as to make it impossible to accomplish anything more in an entire campaigning season. ¹⁷⁶

The huge fortresses, huge armies, and navies of the late ancien régime were only possible because those who commissioned them could take out loans to finance them—loans secured against their revenue for example from taxes. As we saw, such loans were impossible in the pre‐christian and early christian period; and in the late ancien régime they were not available to minor lords. ¹⁷⁷

4.3.2.2 Cultural preconditions: The christian ethos of obedience and obligation

Jean Bodin and Thomas Hobbes explicitly derive the justification of strong ('sovereign’) centralized rule, and of the (near‐)unconditional obedience of the 'subjects’ from the obligation of 'sovereign’ princes to provide security for those subjects: for them, the two are inseparable. Yet in earlier phases of western civilization this type of justification of rule—on the basis of mutual obligation—was absent. Traditional Greek and Roman political culture was predicated on the concept of citizens, free men only beholden to (p. 425 ) their peers. Only tyrants had subjects, but they were by definition illegitimate. Free men could not be subjects, even if they were no citizens, like the metics of Athens. Only slaves were unfree, but this was essentially a category of private law not public law. With the category of subjects essentially absent from political thinking, so was the notion of duties of subjects towards their ruler. Thus, in the post‐Persian Greek world, pólis citizenries on the one hand and kings on the other are always assuring each other of their mutual good will (eúnoia), but on neither side is there any notion of obligation to the other side. ¹⁷⁸ Nor is there any notion of ‘duties’ of 'subjects’ towards the (pre‐christian) Roman ruler, however much that ruler is eulogized in official discourse as all‐powerful.

The only sources that, from a surprisingly early date onwards, very much insist on duties, especially that of obedience, are christian sources. The reason is no doubt that christianity consistently stresses subjection and control. God, but also his representatives—the clergy—are routinely referred to by the term ‘father’, denoting a position of virtually unchallengeable authority in every Mediterranean culture. The term ‘lord’ is routinely applied to god as well as higher clergy. The image of god, or a priest, as a 'shepherd’ is common, as is its converse, the image of the faithful as a ‘flock’ of sheep—animals easily terrified, anything but self‐determined, and weak (note that other herd animals could have been chosen for this simile). Key functionaries of the church bear the title epískopos ‘overseer’, whence ‘bishop’.

True, Zeus is addressed as ‘father’ in Homer, too, if less insistently. Moreover, poimến laố;n 'shepherd of peoples’ is a description routinely applied to kings by Homer. An eastern origin of this simile has been proposed. Being found in Homer, the expression was likewise used by later authors, such as Pindar or the fifth‐century playwrights of Athens. This was an implicit homage to Homer rather than an expression of the prevailing political culture, the more so as Greek society by that time no longer had kings except in Lakedaímôn. Plátôn in the Politikós discusses the simile, though in such an opaque manner that opinions are divided on whether he meant to endorse or reject it. Xenophố;n, however, despite his conservative outlook and his well‐advertised admiration for royalty like the Persian prince Kýros (Kurosh) and Agesílaos of Lakedaímôn, expressly denies the validity of the simile. Tellingly for the political culture of pre‐christian Greek society, he objects that herds of cattle or horses will readily do what the herdsman wants, ‘but men conspire against none sooner than against those whom they perceive to be trying [!] to rule them’. ¹⁷⁹

Unfortunately, virtually nothing survives of the literature on kingship from the post‐Persian Greek realms, with the exception of the so‐called Letter of Aristéas, a second‐century bce Jewish text discussing royal virtues with specific reference to the then king Ptolemy of Egypt, and a number of fragments attributed to ‘Diotogénês’, ‘Sthenídas’, and ‘Ékphantos’ included in the fifth‐century ce anthology of Iôánnês Stobaíos (4.7.61ff.). Whereas the Letter of Aristéas is compatible with the generally libertarian political culture of the pre‐christian Greek world, the fragments strikingly depart from it. Oddly reminiscent in tone of both Thomas Aquinas and Giles of Rome, they depict the king not (p. 426 ) only as practically divine but as indispensable for the social order. ‘Ékphantos’ in particular moreover reflects an image of sacred kingship necessary to a harmonious cosmos and mirroring the superordination of god over the creation, a conception reminiscent of the kind of thinking displayed for example in the preamble of the coronation circular of Henry VII of 1312, or the cosmology of Engelbert of Admont. The status of those fragments is, however, unclear. No other sources mention them, nor can their purported authors be identified with any figures otherwise known; despite the archaic Doric dialect in which they are written they almost certainly date from the post‐Persian, possibly the early christian era. They are not christian themselves, but, like christianity, reflect mid‐eastern influence. ¹⁸⁰

Of course, christianity, through the intermediary of judaism, has mid‐eastern cultural roots, too. Whereas for Xenophố;n it was normal to resist authority, for judaism it was sacred: in the Old Testament man is called on unconditionally to submit to the deity in a manner alien to Greek or Roman religion. In line with this tendency to accept established authority, even Jesus supposedly admonished his listeners to ‘render unto Caesar that which is Caesar's,’ which in the context of the passage specifically refers to taxation. ¹⁸¹ The most famous, and, for the history of christian political culture, momentous call for obedience was formulated around the year 60 by another Jew, Paul:

Let every soul be subject [hypotassésthô/subdita sit] unto the higher powers [exousíais hyperechoúsais/potestatibus sublimioribus]. For there is no power but of God: the powers that be are ordained of God. Whosoever therefore resisteth the power, resisteth the ordinance of God. …Wilt thou then not be afraid of the power? …For he [the ‘ruler’, árchôn/princeps] is the minister [servant: diákonos/minister] of God to thee for good. But if thou do that which is evil, be afraid; for he beareth not the sword in vain …Wherefore ye must needs be subject [dió anánkê hypotássesthai/ideo necessitate subditi estote], not only for wrath, but also for conscience sake. For for this cause pay ye tribute [tax] also …Render therefore to all their dues: tribute to whom tribute is due; custom [toll, customs duties] to whom custom; fear to whom fear; honour to whom honour. ¹⁸²

All this was new to western civilization. Although, in the pre‐christian Graeco‐Roman world, those in power would be seen as enjoying the favour of the gods, they were not regarded as ‘ordained of’, installed by the gods in any institutionalized fashion. One might revere them, but deferring to them was an act of prudence or indeed good will (eúnoia), not a sacred obligation. Not even the ‘emperor cult’ of the Roman empire—another eastern import—actually involved veneration of the living ‘emperor’ as a god, or an obligation to obey him. Rather, what was paid homage to—not, unlike ‘real’ gods, prayed to—was his genius, the Latin word for each individual's personal protective spirit; only after his death was an ‘emperor’ declared to be divine by the senate (an act known as consecratio). The ruler might be addressed as ‘divine’ even while he was still alive (as is Antoninus Pius in the second‐century decree by the Lykian league quoted in Chapter 3 ), ¹⁸³ but this was in the context of a religion that had a more casual attitude (p. 427 ) towards the divine than did christendom. By contrast, the new ethos of obedience insisted on (here and elsewhere) by Paul was meant very seriously, giving the Roman ruler a potentially much stronger position than he held before.

It is true that Roman society had always been somewhat more ready than the Greek world to accept authority, and that it had always indulged in a certain paternalism, manifested notably in the institution of clientship discussed earlier. Indeed, senators were addressed as patres, fathers, and Augustus and his successors were honoured with the title pater patriae, father of the fatherland. As early as the late first century, Domitian insisted on being addressed as dominus, lord, the first Roman ruler to do so, if also, for a long time, the last. The title ‘lord’ was taken up anew only in the third century, somewhat before the adoption of christianity by those in charge of the empire. Absent from the second‐century decree of the Lykian league honouring Antoninus, the title ‘lord’ is used in the—non‐, indeed anti‐christian—third‐century petition from Arýkanda also quoted earlier. ¹⁸⁴ This may be another indication that the same (eastern) cultural exemplars drawn on by Paul were also at work independently of judaism. At any rate, once christianity became the official religion of the empire, obedience, subjection, to its ruler, and generally ‘the powers that be’, at last did become a sacred obligation.

From at least the fifth century onwards, the ruler of the empire would routinely claim to owe his office to god: a gesture of humility at the same time as it bestowed a religious legitimacy unknown in the pre‐christian empire. Thus, the western augustus Marcianus (in power 450–7) wrote to pope Leo I that he had gained his supreme position through god's providence as well as election by the senate and the whole army. ¹⁸⁵ Iustinus (in power 518–27) similarly wrote to pope Hormisdas that he reigned by the favour of the Holy Trinity, as well as election by the dignitaries of the palace, by the senate, and by the whole army. ¹⁸⁶ Justinian (in power 527–65) credits god as the one who put him in power over the empire, entrusted to him by the ruler of the heavens (Deo auctore nostrum gubernantes imperium, quod nobis a caelesti maiestate traditum est), and protests that rather than rely on his army or his reason he puts all his hope in the providence of the Holy Trinity (ut …omnem spem ad solam referamus summae providentiam trinitatis). ¹⁸⁷

In all those instances, god is not yet the sole source of legitimacy mentioned, as he would later become when kings, following the example of the Carolingians, would simply claim to rule ‘by god's grace’, dei gratia. But already god was always named first, and the simplification that consisted in naming only god was merely a matter of logic and time. In parallel with this development, the Roman ruler rose above the law. In his eulogy of Trajan, addressed to the Roman senate in the year 100, the younger Pliny still insisted that ‘the ruler is not above the law, but the law above the ruler’. ¹⁸⁸ By contrast, the Corpus iuris promulgated by Justinian states that the ruler is not bound by the law, princeps legibus solutus, a formulation credited in the text to Ulpian, a legal scholar active around the turn of the third century. Whatever standing this opinion had at that time, (p. 428 ) its inclusion in the Corpus iuris made it part of official Roman law, with its enormous influence not just on Latin christendom, but even more so post‐Reformation Europe. ¹⁸⁹

On the other hand, the new concept of a sacred obligation of obedience was matched by the equally novel notion that the ruler himself had duties. He was himself subject, and thus answerable for his acts, to god. Paul, in the passage quoted, calls him a 'servant’ of god, which would have been inconceivable in a non‐christian Graeco‐Roman text. No more than a pre‐christian Greek or Roman ruler would be seen as installed by the gods would he be considered their servant. He was as little beholden to the gods (although of course he was well‐advised to stay on good terms with them) as he was to those over whom he exercised power.

Seneca, in exhorting Nero to refrain from cruelty in his treatment of those under his sway, invokes no moral argument but the idea that a cruel ruler will gain a bad reputation passed on even to posterity. He compares the position of the ruler with that of the father with regard to his children (in traditional Roman law, a father had power of life and death over his offspring), the teacher with regard to his pupils, the commander (tribunus) with regard to his soldiers, indeed, if somewhat obliquely, the master with regard to his slaves. (Seneca is of course exaggerating the power that this implicitly ascribes to his addressee, whom is seeking to flatter.) All those instances have in common their one‐sidedness, the absence of any mutual obligation—certainly at that time. In every case, those in the inferior position could literally only hope for the best, for generosity on the part of the superior, which, especially in the case of a ruler, would be celebrated the more effusively as it was entirely voluntary. But neither, for Seneca was obedience an absolute imperative. He notes that a cruel father, teacher, or commander will often find his behaviour to produce undesired results: thus, a cruel commander ‘will produce deserters, whom yet no one will blame [desertores faciet, quibus tamen ignoscitur]’. ¹⁹⁰

Compare this with the admonition addressed to the Frankish king Charles I by his English adviser Cathwulf in a famous letter of about 775:

Therefore, my king, always be mindful of god who is your king, with fear and loving, because you have been set as his vice‐regent over all his limbs [the members of the christian people, regarded as the mystical body of Christ], to guard and direct them, and to give account on the day of judgment, even for yourself. ¹⁹¹

The idea that the gods might call Nero to account is present in Seneca, too (cf. the passage quoted in Chapter 2 , n. 246), but the emphasis is different: it is treated as a hypothetical possibility, and one moreover that would simply give Nero another occasion to shine. If Seneca, whose real feelings for his one‐time pupil were no doubt mixed at best, purports to treat Nero with unalloyed deference, it is because he had no demands to make. Cathwulf, by contrast, feels free, in the name of religion, to call on the powerful king of the Franks to be both responsible and humble, reminding the king that he will have to account for his reign and that how he will be judged is no foregone conclusion. (p. 429 ) Three centuries later, in a letter already quoted, pope Gregory VII would make that point quite stridently:

Kings and other princes ought to be afraid lest they be made to suffer the eternal fires of hell to a greater extent in proportion as in this life they enjoy preferential treatment over others …For they shall have to give account before god for all those that were subject to their rule. ¹⁹²

The notion of service is a corollary of the christian emphasis on obedience and finds expression already in the gospel:

whosoever [Jesus says to his disciples] will be great among you, let him be your minister [=(free) servant: diákonos/minister]; And whosoever will be chief among you, let him be your servant [doúlos/servus: this typically means 'slave’]: Even as the Son of man [=Jesus] came not to be ministered unto [=to be served, diakonêthếnai/ministrari], but to minister [=to serve, diakonếsai/ministrare], and to give his life a ransom for many. ¹⁹³

The injunction was, to some extent, taken seriously by those in power. Since Gregory I (in office 590–604), the pope has referred to himself as servus servorum dei, servant of the servants of god. ¹⁹⁴ Coins from the reign of Justinian II (basileús 685–95 and 705–11) show him standing with a large cross in his hand and the legend servus Christi, servant of Christ. ¹⁹⁵ The emperor Otto III (late tenth century) included in his title the words servus apostolorum, servant of the apostles, and Henry II in the early eleventh century called himself servus servorum Christi, servant of the servants of Christ. ¹⁹⁶ Iôánnês Zônáras, in the twelfth century, exhorts the ruler of Constantinople to be a steward, oikonómos, not a despótês, a lord or master, with regard to the commonwealth (politeía), ¹⁹⁷ and Nicholas of Kues similarly insists that the emperor is not a lord or master but a steward, non dominus sed administrator. ¹⁹⁸ Engelbert, as we saw, thought that ‘in the natural order of things the king exists for the sake of the kingdom and not the other way around’. ¹⁹⁹ Similarly, for Dante

The citizens do not exist for the sake of the consuls and the people do not exist for the sake of the king, but rather the consuls exist for the sake of the citizens and the king for the sake of the people …Even though the consul or the king are lords of the rest with respect to the path, they are servants of the rest with respect to the goal, but especially the monarcha [emperor], whom we obviously must consider the servant of all [minister omnium]. ²⁰⁰

This is echoed in famous formulations by Frederic II of Prussia (reigns 1740–84): ‘the sovereign, far from being the absolute master of the peoples under his domination, is really (p. 430 ) only their chief servant’; ²⁰¹ or: ‘A prince is the first servant and the first magistrate of the state; he is answerable to it for the use to which he puts the taxes.’ ²⁰² This whole theme would have been alien to the political culture of the pre‐christian Mediterranean world, where those in power owed nothing to those over whom they had power even if they were expected to extend their generosity towards them. In this respect, the triumph of christianity marks a profound change, which proved durable even when the attraction and power of the church began to wane again—as Frederic, who was no friend of the church, demonstrates.

Christian and post‐christian rule was thus necessarily conditional. A prince now had subjects, but he also had duties towards them. Something was expected of him; what, precisely, remained open to definition. Thomas Aquinas and Giles of Rome, as we saw, provided some answers: for them, a king had to maintain ‘peace’, or unity, among his subjects, incite them to be virtuous, punish wrongdoers, and direct them towards the life eternal. Nicholas of Kues, who cites them, ²⁰³ imitates them in this. Yet Nicholas, amalgamating the stress on popular participation that he found in Marsilius with the growing emphasis on strong princely governance, also foreshadows two other important aspects of post‐Reformation political thought: the notion of the social contract—for Nicholas the ruler is ruler only by virtue of the consent of the ruled—and the expectation that a strong ruler will, above all, provide security. Although the term as such is not yet prominent in his text, the notion is implicit in his criticism of the situation in Germany, where according to him a weak central power meant that no one was safe (nullus tutus).

In Latin christendom, rule was conditional also in the sense that subjects retained the right to question decisions or demands by the ruler, and to disobey if they held them to be illegitimate. This had something to do with the difficulty faced by those in power to enforce decisions on a supralocal scale, a difficulty that diminished with the progress of monetization and the greater facility with which a ruler could act, including supralocally, through paid agents, armed (i.e. soldiers) or not (i.e. administrators or judges). The growing potential power of the ruler provoked two types of reaction: enthusiasm as well as vigilance. Some felt that greater power necessitated greater control. As mentioned, representative assemblies of the ‘estates’ of a realm or principality sprang up or were consolidated everywhere in Latin christendom in the fifteenth century, and the conciliar movement in a sense is simply the ecclesiastical equivalent of this phenomenon. Since the increasing potential power of princes was linked to the progress of monetization, tellingly what assemblies of the estates sought to control above all was taxation. Others, however, impatient with attempts to hem the ruler in, felt that the time was ripe for ‘real’ governance by a powerful ruler, which they hoped would prove the cure for all sorts of social ills. Nicholas of Kues, as we saw, wavered between those two impulses, while Enea Silvio Piccolomini came down firmly on the side of complete subjection and obedience.

So, after the Reformation, did the most influential theorists of strong princely governance, Jean Bodin and Thomas Hobbes. Stressing as they did the fullness of the power of the ruler, they were also at pains to make clear what was expected of that ruler (p. 431 ) in return. Here the concept of security became central. However, the opposite strain, that of the need for the subjects to keep their ruler in check, continued into the post‐Reformation period, too, being represented for example by Johannes Althusius.

4.3.3.1 ‘Absolutists’ and ‘monarchomachs’: Jean Bodin and Johannes Althusius

In his Coutumes de Beauvaisis of 1283, a codification of the customary law of the Beauvais region, Philippe de Beaumanoir explains that

Since in this book in several places we speak of the souverain and of what he may or ought to do, but do not mention counts or dukes, some might think that reference is to the king. But whenever the king is not mentioned, we refer to those holding a fief, for every holder of a fief is souverain within it. However, the king is souverain above them all [par dessus tous], and by rights has the general care of his kingdom, whereby for the common weal he may make as many statutes as he likes, and what he lays down must be observed. ²⁰⁴

The passage shows that, on the one hand, there was no royal monopoly on political power, and on the other that nevertheless the king had a special position. Souveraineté was more ‘naturally’ attributed to him than to other lords, whose similar, if geographically more restricted competence Philippe finds it necessary to remind readers of. ²⁰⁵

The word souveraineté is in fact derived from the Latin comparative superanus, 'superordinate, superior, higher‐ranking’, and not from the superlative supremus, 'supreme, highest’. (In the Latin of Latin christendom, the adjective superanus tends to supersede the ‘classical’ forms superus and supernus.) From the fifteenth century onwards, high‐ranking law courts in Paris and other French cities came to be known as cours souveraines—an expression of course of their superordination over other courts and not of any 'sovereignty’ in the more recent sense. It was only in 1661, when Louis XIV took power, that he had the adjective changed from 'sovereign’ to 'superior’ (cours supérieures—but historians of eighteenth‐century France routinely refer to the cours souveraines even if they mention the 1661 measure, so perhaps the new appellation (p. 432 ) did not prevail). Earlier, in his highly influential Six Books on the Commonwealth first published in 1576, the legal scholar Jean Bodin (1530–96) had taken the step that we find foreshadowed in Philippe de Beaumanoir, redefining souveraineté as absolute rather than relative superordination. Whereas Philippe denied that only the king had sovereignty, Bodin asserts precisely that.

Bodin starts from the position outlined in the preceding section. Rule for him means exercising an office and serving a function, and being answerable to those for whose sake that office is held and the function fulfilled. There is a mutual obligation between the ruler and the ruled: ‘The prince’, Bodin writes, ‘is obliged [obligé] to preserve [maintenir], by military force and the force of the laws, the safety [seureté] of his subjects [subiects] with respect to their persons, belongings, and families; and the subjects, by reciprocal obligation [par obligation reciproque], owe their prince fealty, subjection, obedience, aid, and assistance [foy, obeissance, subiection, ayde, & secours].’ The relationship between the ruler and the ruled is still conceived here on the model of the ‘feudal’ order with its emphasis on reciprocal loyalty. But, unlike the social paradigm of Latin christendom, here there is only one lord, all others are 'subjects’. ²⁰⁶

It is true that the concept of sovereignty even as redefined by Bodin still falls short of giving the ruler certain rights that nowadays are taken for granted as being vested in the 'state’ or ‘government’. Thus, Bodin denies that the ruler may dispose of the property of his subjects—probably because the right of property was considered by Bodin to be part of natural law, which even the 'sovereign’ could not modify. Bodin is even opposed to the levying of taxes by the ruler, presumably because this, too, meant encroaching on the property of the subjects. Bodin holds that the king should properly be satisfied with the income of what belongs to the crown, that is, the royal demesne. He insists that even in a public emergency, such as in wartime, extra funds should be raised through loans rather than taxes, presumably because he feared that once granted such taxes would be difficult to abolish again. ²⁰⁷ Moreover, there were, for Bodin as for other French lawyers of the period, certain fundamental laws of the realm that the ruler had no authority to change. In the French case, this was true notably of the order of succession to which the ruler himself owed his office. In the French text, Bodin refers to such fundamental laws as loix royales ‘royal laws’, but in his own Latin translation of the work he calls them leges imperii ‘laws of the empire’—in line with the tendency at that time to equate regnum and imperium, that is, to treat every kingdom as a legal equivalent of the (Roman) empire. ²⁰⁸

Employing ‘imperial’ terminology, French lawyers thus spoke of the French kingdom as a corpus mysticum (the mystic body of Christ: this originally meant christendom, which, however, thinkers like Nicholas of Kues or Enea Silvio Piccolomini equated with the empire), or of the Parlement de Paris as an analogue of the Roman senate. ²⁰⁹ The Tudor kings and queens of England likewise display a certain fondness for such terminology. Henry VIII repeatedly qualifies his crown as ‘imperial’ in his official utterances: ‘[the] crown imperial of this our realm’ (in the proclamation rejecting papal authority over the church of England, 1535), ‘our imperial crown of our realm of England’ (in the proclamation announcing the adoption by Henry of the title ‘king of Ireland’, 1542). On (p. 433 ) their accession, Edward VI in 1547 was proclaimed heir to the ‘crown imperial of this realm, and other his realms’, and Mary I in 1553 was proclaimed heiress to the ‘crown imperial of the realms of England and Ireland, with the title of France’. When Elizabeth I was proclaimed in 1558, the document only spoke of ‘the crown of the foresaid kingdoms of England, France, and Ireland’. But a few weeks later another proclamation specified her exact title as being ‘the most high and mighty Princess, our dread sovereign Lady Elizabeth, by the grace of God Queen of England, France, and Ireland, defender of the true, ancient, and Catholic faith, most worthy Empress from the Orkney Isles to the Mountains Pyrenée’. This kind of grandiloquent, and often very long, title is typical of the post‐Reformation period but not of pre‐Reformation Latin christendom, and again demonstrates the heightened profile of the ruler in the 'social episteme’ of the period. ²¹⁰

In a book published in 1566, Bodin himself had concurred in the opinion then prevailing among students of French law that the power of the French crown was limited, especially by the prerogatives of the Paris parlement, its counterparts in the French provinces, and the assemblies of the estates, and that the king could not legislate on his own. ²¹¹ By contrast, in De la république Bodin holds that the king of France alone held supreme power in the kingdom and was the sole possessor of what he calls rights of sovereignty, or in Latin iura maiestatis. ²¹² The fact that despite many intrinsic contradictions and logical weaknesses this book caused so much interest—it was reprinted countless times, as well as quickly translated into Latin (1586), Italian (1588), Spanish (1590), German (1592), and English (1606)—shows that Bodin had hit a nerve. Ultimately, the controversy that it helped to fuel oscillated between poles already represented by Nicholas of Kues and Enea Silvio Piccolomini (and indeed by Bodin himself, in 1566 and 1576 respectively): is the possessor of supreme power also, to some extent, controlled by the subjects, or only they by him? One of those who explicitly took issue with Bodin was Johannes Althusius (1563?–1638), professor at a Calvinist college at Herborn (in northern Hesse), then 'syndic’ of the town of Emden in East Frisia, in a Latin work entitled Politics, systematically discussed and illustrated with sacred and profane examples.

Although his book did well (four editions in his lifetime), Althusius never quite achieved the popularity, or lasting fame, of Bodin, and as far as I am aware there is at the time of this writing still no full translation of the Politics in any other European language. ²¹³ Althusius shares this fate of relative oblivion with fellow ‘monarchomachs’, (p. 434 ) or ‘fighters against the monarch’, as the Englishman William Barclay termed them in 1600 (Barclay himself belonged to the opposite camp, the ‘absolutists’ or advocates of ‘absolute’ monarchy, but this is not a contemporary coinage: the word ‘absolutism’ only appeared in the nineteenth century). ²¹⁴ Other such monarchomachs included Frenchmen like Theodore Beza (Théodore de Bèze) and the author or authors of the anonymous Vindiciae contra tyrannos (‘A defence against tyrants’) of 1579, many of whose arguments are taken up by Althusius, or the two Scotsmen George Buchanan and John Knox. All of them adhered to calvinist protestantism, a creed particularly prone to preaching resistance to the crown for conscience's sake, and an inspiration also for the English ‘puritans’, who played a major role in the temporary overthrow of the British monarchy in the mid‐seventeenth century. None of them, however, offers a global, comprehensive theory of society and politics in the manner of Althusius, who therefore seems best suited for a comparison of this general position with the ‘absolutism’ of Bodin and Hobbes.

Unfortunately, for the monarchomachs, the prevailing 'social episteme’, or ideological fashion, favoured a strong central power, and thus authors who supported this idea. That the two opposing camps actually shared many premisses is shown by the fact that their respective adherents might quite easily switch between them. We saw that Bodin in 1566 argued that the French crown shared its power with other political agents within the kingdom, only to contend that it did not in 1576. Likewise, in a textbook on Roman law the young Althusius had affirmed in the 1580s that the ruler was suis et alterius legibus solutus, not bound by any laws made by himself or by others; that the ruler had plenum et absolutum imperium, full, unrestricted power of command; and even that he was entitled to levy taxes—which Bodin, whom Althusius in his treatise cites frequently, disapproved of. ²¹⁵ Althusius, too, later changed his tune completely.

As 'syndic’ Althusius was really ‘only’ legal adviser and representative of the town council of Emden. In reality, however, he acted very much like the leader of the citizenry, encouraging their resistance to their nominal overlord, the count of Frisia. The town, on the mouth of the river Ems, is situated just east of the border dividing West Frisia, one of the provinces of the Low Countries and now part of the Kingdom of the Netherlands, and the county of East Frisia, now part of the Federal Republic of Germany. Following its near‐total destruction in the air raids of the Second World War, little now reminds the visitor to this town of its wealth and self‐confidence in Althusius’ day. The rebellion of what became the ‘United Provinces’, the calvinist‐dominated northern provinces of the Low Countries, against the Spanish crown from 1566 onwards proved a boon for Emden. Its port profited greatly from being safe from the war going on further west, and its population was swelled by calvinist merchants and other refugees. Unlike calvinist West Frisia, the county of (East) Frisia was lutheran. But with its size and prosperity thus enhanced, and its religious orientation altered, the town in 1595 officially abandoned its lutheran affiliation in favour of calvinism, and in the treaty of Delfzijl—endorsed by the emperor Rudolf—forced the count to give up much of his prerogatives over the town, though his suzerainty was confirmed.

(p. 435 ) Friction between the town and the count continued, however. In 1604, the town council dismissed its syndic as being too sympathetic to the count and offered the post to Althusius, recommended by his Politica, first published the year before and very much in favour of restricting the power of the ruler. Althusius accepted, a decision no doubt facilitated by the ‘rather extraordinary salary’ (as one contemporary source puts it) that the town was willing to pay him and which later was repeatedly increased further; Althusius remained in office until his death in 1638. Under his leadership, the town in 1608 again rebelled against the count, who saw his residence and archive at Aurich pillaged by troops maintained by the estates of the county supposedly for its defence but manipulated by Althusius; in 1618 the town forcibly detained the count, who had come there for talks, for several weeks. When the count died, Althusius insisted that the town should no longer swear ‘obedience’ to his successors, but only ‘loyalty’. Althusius pursued a policy of strengthening the estates of the county generally, against the count, but did not hesitate to put the interest of Emden above that of its fellow estates. Although Emden had been equipped with modern fortifications at the expense of the estates, Althusius after the outbreak of the Thirty Years War in 1618 refused to contribute to the defence of the county as a whole, and with his pro‐Dutch attitude undermined efforts to keep the county neutral and out of harm's way. During the Emden years, Althusius meanwhile continued to work on the Politica, subjecting it to many modifications. ²¹⁶

The Althusius of the Politica rejects Bodin's opinion that sovereignty is vested in the ruler alone. Already in the preface, he formulates the opposite view:

I have attributed the so‐called rights of sovereignty [jura majestatis] not to the supreme magistrate [summus magistratus] but to the commonwealth or in other words the universal association [respublica vel consociatio universalis]. Most legal and political authors attribute them to the prince [princeps] and supreme magistrate alone. …By contrast, I, along with a few others, maintain that they belong to the symbiotic body of the universal association [corpus symbioticum consociationis universalis]. …I recognize the prince as the administrator, trustee, handler of the rights of sovereignty [administrator, procurator, gubernator jurium majestatis], but as the owner [proprietarius] and usufructuary [usufructuarius] of sovereignty [majestas] none other than the people as a whole [populus universus]. …For those rights of sovereignty establish and preserve [constituunt et conservant] the universal association. And as they have their origin [inceperunt] in the people, or the members of the commonwealth or realm [populus, seu membra Reipublicae vel regni], so also can they only exist in them, and be preserved by them [non nisi in illis consistere possunt, et ab illis conservari]. Likewise, their administration [administratio], entrusted by the people to a single mortal, the prince or supreme magistrate, reverts to the people …when he dies or is deprived of his office [exauthoratus]. ²¹⁷

The terminology—majestas, jura majestatis—is Bodin's: even in standing the Frenchman on his head Althusius would be unthinkable without him.

If Bodin's theory put the emphasis on the ruler, Althusius stresses the primacy of the people:

(p. 436 )

The commonwealth or realm does not exist for the sake of the king, but the king and every other magistrate exists for the sake of the realm and of the polity [politia]. For the people, prior, in the nature of things and in time, to its leaders, is more powerful than they, and above them [Nam populus, natura et tempore prior, potior, et superior est suis gubernatoribus]. ²¹⁸

It follows that a people can exist without a magistrate, but a magistrate cannot exist without a people. ²¹⁹

Engelbert and Dante had said much the same, but did not go quite so far: they did not place the people before and above the king, at least not explicitly, nor did they assert that the people could do without him, or leaders in general.

Not least because Althusius did not regard the ruler as constitutive for the commonwealth, he was at liberty, unlike Bodin and Hobbes, not to regard that commonwealth itself as monolithic—as indeed no large realm of his day was. Althusius treats the commonwealth as a hierarchy of communities much like the scholastics, but in a more elaborate manner. It is the totality of those communities that for Althusius represents ‘the people’. ‘Private associations’, in this system, include families and kinship groups (familiae et cognationes) established by marriage or blood ties, and ‘minor corporations’ (collegia minora—such as craftsmen's guilds, but also associations of clerics, judges, and the like). ‘Public associations’ include municipalities (universitates), defined as ‘associations, established by means of laws, of a number of married couples, families, and corporations and inhabiting the same place’ ²²⁰ and taking the form of villages (vici), districts (pagi), towns (oppida), and cities (urbes); and provinces (or lands: provinciae), represented by their estates (ordines). The estates are made up of ‘great corporations’ (collegia maiora): clergy, gentry, towns, and peasants. Althusius regrets that peasants were not everywhere represented in assemblies of the estates (they were in Frisia). ²²¹

For Bodin, the commonwealth is made up of households (mesnages, familiae). By contrast, Althusius emphasizes that it is towns and lands that make up the commonwealth, not households or local corporations, just as a ship is not formed by planks, nails, and dowels, but by the prow, stern, and keel. ²²² Bodin, too, considered corporations and estates useful and advocated their consultation. ²²³ But essentially, for him, there is the ruler on the one hand, the subjects on the other. For Althusius, ‘politics’ is not primarily about subjection or the exercise of power.

Politics is the art of associating people for the purpose of establishing, cultivating, and preserving a communal life among them. That is why it is called symbiotics [i.e. the art or science of living together; I have not been able to find out what other authors, if any, employ this term apart from Althusius.] The object of politics is the association by which, through an explicit or tacit agreement, those living together [symbiotici] enter into a mutual obligation to share whatever is useful or necessary for the communal life and fellowship. ²²⁴

(p. 437 ) The communities, on every level, of ‘those living together’ are based on such an ‘explicit or tacit agreement’. Althusius is evidently inspired by a definition of Cicero already cited, and which he, too, adduces: Est igitur res publica res populi, populus autem …coetus multitudinis iuris consensu et utilitatis communione sociatus, ‘The commonwealth’ [literally ‘the public cause’], then, is the cause of the people, and the people …is a group associated by consensus on what is the law, and ties of mutual interest.’ As I have pointed out earlier, this formulation, often taken as a definition of the 'state’, omits any mention of a governing agency, let alone of subordination to it. ²²⁵

Why certain towns and countries should be part of a given ‘universal association’—the expression that Althusius uses as the generic term to designate the greatest, and fully autonomous, political unit—but not others seems not to be explained in the book, except indirectly in that Althusius considers the consociatio or association to be voluntary at every level. This is true all the way down to the smallest and most basic units, and despite the fact that these are described as ‘natural’: even the domestica et naturalis consociatio of families and kinship groups consists of individuals bound together, and acknowledging a single leader, ‘by their own consent [consensu suo]’. ²²⁶ Similarly, Althusius says of the consociatio universalis that it is ‘a people, merged into one body by the consent of several symbiotic associations and particular corporations, or of the bodies of several associations, and gathered under one law’. ²²⁷ Membership in a ‘universal association’, then, is in effect based on a right of self‐determination exercised separately by each of the subdivisions, or smaller associations, forming part of a commonwealth.

Decision‐making for Althusius is mainly in the hands of assemblies recruited from among the members of the various associations—the councils (senatus) of the towns, the estates of the provinciae, the estates of the realm in its entirety, with the latter described as as ‘the epitome [epitome] of the polity or realm’. ²²⁸ Concerning the ruler, Althusius pushes to its extreme the notion that, a mere 'steward’ and 'servant’, he has duties to fulfil—failing which he can be dismissed: a corollary that earlier authors had been reluctant to admit.

The people has entrusted itself to the governance of the king on certain conditions and terms, retaining, in certain conditions, power and authority [potestas et autoritas] over the king should he prove unworthy. His leadership and stewardship is not unrestricted, absolute, and completely free [Directio et administratio haec, non est plena absoluta et liberrima] …but aimed at their welfare and circumscribed and limited by certain laws. …Individual subjects are under the king, but in their entirety they are above the king. …If he exercises his stewardship properly, he is paramount [praeest] with regard to the individuals to the extent that he is the executor, guardian, and servant of the law [quatenus legis est exsecutor, custos et minister]. …Hence, when he commands something that runs counter to what the law lays down, he falls under the (p. 438 ) power of the law and loses his superior status; in that case he becomes subject to those who carry out the law [Ergo contra legis praescriptum imperans, legi fit obnoxius, et superior esse desinit; quo casu incipit subesse legis exsecutoribus]. …As a result …he ceases to be king and a public person and becomes a private person, and thus if he proceeds and acts with evident injustice anyone is free to resist him. ²²⁹

By contrast, for Bodin the ruler was above the law, and he denies that active resistance even to a tyrant was permissible; he does concede a right of disobedience if what the king commands runs counter to divine or natural law. ²³⁰

What are the sovereign rights (iura maiestatis)? According to Althusius, generally they concern the implementation of what he calls the ius commune or common law, which for him consists in basic, natural law norms that he sees exemplified in the Ten Commandments. Apart from that, a given realm will have need of specific norms, tailored to its particular requirements, in such domains as trade, money, language, public offices, the bestowal of privileges, defence, the upkeep of common institutions such as a common treasury, arsenals, archives, and the like. ²³¹

If I am not mistaken, Althusius never spells out why the ruler is necessary at all, or if there is anything that should properly be left to his discretion. Whereas Nicholas of Kues, Bodin, or Hobbes all thought that the ruler was indispensable to ensure the safety, the protection of the subjects, Althusius, clearly in reaction to a climate of opinion eager to empower the ruler, seems almost exclusively concerned with protecting the people from their ruler. Far from pining for a strong ruler, Althusius actually posits that he should be kept weak. What power the ruler has is determined by the contract by which the people give him his mandate (pactum seu contractus mandati). ²³²

The supreme magistrate possesses as many rights [tantum juris] as are expressly awarded to him by the associated bodies, or members of the realm, and what is not awarded to him has to be considered as remaining with the people, or universal association; for this is in the nature of the contractus mandati. The less power is given to those in command [qui imperant], the more long‐lived and stable the realm [imperium]. …Absolute power, or fullness of power, as it is called, cannot be given to the supreme magistrate [Absoluta potestas, seu plenitudo potestatis, quam vocant, summo magistratui dari non potest]. ²³³

At his inauguration, the ruler had to swear to respect the laws of the realm: this was the practice everywhere in Europe, as Althusius is able to illustrate with abundant examples. The French king, too, did so at his coronation, posing a problem for Bodin: how could the king be above the law if he had to swear to uphold it? Somewhat lamely, Bodin argues that he was not actually bound by his oath, only expected not to alter the laws in question except for a weighty reason. ²³⁴

For Bodin (at least the Bodin of De la république), the sovereign is sole legislator. Althusius has no patience with this.

Nowadays general laws are made by assemblies. For it would be a mark of reckless and foolish arrogance for one man, or a few men, to consider themselves capable of giving adequate laws to some nation [gens], without its assent and the combined advice and opinion of many. ²³⁵

(p. 439 ) What Althusius depicts is not so much a programme as a (somewhat pointed) description of what he saw around him, in Frisia, in the Netherlands, in the Holy Roman Empire, in Europe as a whole: everywhere estates and their assemblies were powerful, nowhere was the ruler a real 'sovereign’. France is a case in point: when Bodin wrote De la république, the country was shaken by religious turmoil and the crown was weak. Likewise, Hobbes published his Leviathan at a time when the English monarchy had in fact just been abolished, a result, not least, of the kind of ideas propagated by people like Althusius.

4.3.3.2 Hobbes and the beginnings of IR theory

The sovereign as a prerequisite of society: a break with tradition. Thomas Hobbes (1588–1679) is the most radical exponent of the intellectual current favouring a strong, even omnipotent ruler. Like Bodin, he wrote his masterpiece Leviathan, or The Matter, Forme & Power of a Common‐Wealth Ecclesiasticall and Civill against the background of political turmoil experienced first‐hand: when the book was first published in 1651, two years after the execution of king Charles I of England and Scotland, Hobbes was living in exile in Paris. Hobbes explicitly states that the work was ‘occasioned by the disorders of the present time’. ²³⁶

Among the writers so far encountered in our overview of the western tradition, Hobbes stands out because of his explicit refusal to accept that tradition. His political thinking would be inconceivable without the christian ethos of obedience and obligation, or the redefinition of sovereignty popularized by Jean Bodin. But as Hobbes saw it, all western political thought prior to his own was pretty much useless. Whereas Bodin and Althusius are full of references to other authors, such references are largely absent from Hobbes. Only the very greatest, and ancient, western thinkers are deemed worthy of being named—albeit, more often than not, only to be trashed. Aristotle is singled out for the most acerbic criticism.

I beleeve that scarce any thing can be more absurdly said in naturall Philosophy, than that which now is called Aristotles Metaphysiques; nor more repugnant to Government, than much of that hee hath said in his Politiques; nor more ignorantly, than a great part of his Ethiques. ²³⁷

Hobbes in fact repudiates Graeco‐Roman political thought generally.

In these westerne parts of the world, we are made to receive our opinions concerning the Institution, and Rights of Common‐wealths, from Aristotle, Cicero, and other men, Greeks and Romanes, that living under Popular [=democratic] States, derived those Rights, not from the Principles of Nature, but transcribed them into their books, out of the Practice of their own Common‐wealths, which were Popular. …And by reading of these Greek, and Latine Authors, men from their childhood have gotten a habit (under a false shew of Liberty,) of favouring tumults, and of licentious controlling the actions of their Soveraigns; and again of controlling (p. 440 ) those controllers, with the effusion of so much blood; as I think I may truly say, there was never any thing so deerly bought, as these Western parts have bought the learning of the Greek and Latine tongues. ²³⁸

The great problem with pre‐christian political thinkers was their rejection of dominion, of strong rule, particularly by a single person, which had exercised a deplorable influence on all later western thinking: ‘as to Rebellion in particular against Monarchy; one of the most frequent causes of it, is the Reading of the books of Policy, and Histories of the antient Greeks, and Romans’ ²³⁹ —even though, actually, more recent calvinist writers, not of course mentioned by Hobbes, were perhaps more to blame. Western thinking, for Hobbes, being thus hopelessly vitiated, it was necessary to start afresh, which is what Hobbes proposes to do. Salvation from all sorts of social problems lay in strong, unitary rule, preferably by one man. As we have seen, this was hardly a new idea. But Hobbes fails to acknowledge any positive contribution to his own theory by earlier authors and instead implicitly dismisses them all for not having been sufficiently radical. One had to be systematic and rigorous in developing a theory of politics, not intuitive and pragmatic—‘unlesse we shall think there needs no method in the study of the Politiques, (as there does in the study of Geometry,) but onely to be lookers on; which is not so. For the Politiques is the harder study of the two.’ ²⁴⁰ Not for the last time in the academic treatment of politics (or international relations), what Hobbes was dreaming of was social science as ‘hard science’: ‘The skill of making, and maintaining Common‐wealths, consisteth in certain Rules, as doth Arithmetique and Geometry; not …in Practise onely.’ ²⁴¹

If there really were such rules of politics, why had no one yet found them? Obviously, the Greeks and Romans had something to answer for, but in dealing with this anticipated objection Hobbes here puts forward a different argument: that, quite simply, someone had to be the first to find those rules. ‘Time, and Industry, produce every day new knowledge.’ ²⁴² In reality, what Hobbes offers is not new knowledge but skilful advocacy, under cover of scientific objectivity, of an ideology that (albeit in less radical form) many people of his time adhered to anyway and would have adhered to without him. He presents an impressively coherent body of thought, but the simple premisses with which he begins are wilfully selective and intrinsically questionable—for example: ‘in the first place, I put for a generall inclination of all mankind, a perpetuall and restlesse desire for Power after power, that ceaseth onely in Death.’ ²⁴³ What Hobbes presents as deductive reasoning is actually inductive: it was the message itself, a personal, subjective, political choice on issues hotly debated in his day, that was Hobbes’ real starting point.

Hobbes’ sovereign is not bound by human law of any kind. As a result, it is not possible to enter into a treaty with him: he cannot be bound by it. The social contract, then, that had been posited at least since Nicholas of Kues, for Hobbes cannot be concluded between the sovereign and his subjects, but only among the subjects themselves: they agree among themselves to give up all power to the sovereign. But Hobbes is splitting (p. 441 ) hairs. Although he claims that the sovereign cannot be bound by this treaty, Hobbes nevertheless admits that, along with the power, it bestows on him an ‘end’, a function—‘The Office of the Soveraign …consisteth in the end, for which he was trusted with the Soveraign Power, namely the procuration of the safety of the people’ ²⁴⁴ —and that if he fails to fulfil this function he loses his position:

The Obligation of Subjects to the Soveraign, is understood to last as long, and no longer, than the power lasteth, by which he is able to protect them. For the right men have by Nature to protect themselves, when none else can protect them, can by no Covenant be relinquished. ²⁴⁵

It is difficult to see how this is different from a treaty whereby the sovereign promises protection in return for power.

In any case, even the treaty between the subjects is actually fictitious. Hobbes infers from his analysis of human nature that autonomous individuals would have to make such a treaty to protect themselves from each other, and further, that the sovereign created by this treaty would then necessarily have the right to treat individuals not parties to the treaty as enemies and force them into subjection. This being a logical necessity, the sovereign is legitimate whether or not such a treaty was ever concluded in fact; for the same reason there is no need ever to renew it. Hobbes actually doubts that any sovereign was ever created from voluntary subjection rather than usurpation: ‘there is scarce a Commonwealth in the world, whose beginnings can in conscience be justified.’ ²⁴⁶ Put the other way around, as long as the sovereign protects the subjects from each other and from outsiders, how he obtained his position is of no importance. Although Hobbes is often described as a ‘contractarian’, a theorist of the social contract, the treaty on which the position of the sovereign is supposedly founded thus turns out to be a redundant element of his thinking.

Hobbes famously posits that there can be no social life, material well‐being, or culture without subjection to, and coercion by, a sovereign:

during the time men live without a common Power to keep them all in awe, they are in that condition which is called Warre; and such a warre, as is of every man, against every man. …In such condition, there is no place for Industry, because the fruit thereof is uncertain: and consequently no Culture of the Earth; …no Arts, no Letters, no Society; and which is worst of all, continuall feare, and danger of violent death; And the life of man, solitary, poore, nasty, brutish, and short. ²⁴⁷

This has long become a familiar passage; but when it was new it shocked readers brought up on the notion propagated by Aristotle that man was by nature a social being, politikón zố; _i on.

Hobbes compares the sovereign to the biblical Leviathan, a mythical creature from whose description in the Book of Job he quotes two verses (41.33–4): ‘There is nothing on earth, to be compared with him. He is made so as not to be afraid. Hee seeth every high thing below him; and is King of all the children of pride.’ ²⁴⁸ The rest of the description, (p. 442 ) not quoted by Hobbes, is quite forbidding. Oddly, too, the prophet Isaiah (27.1) predicts that ‘the lord with his sore and great and strong sword shall punish leviathan the piercing serpent, even leviathan that crooked serpent,’ and Psalm 74 praises god for delivery from this monster (verse 14: ‘Thou brakest the heads of leviathan in pieces’).

But, if an all‐powerful ruler was so desirable, why had people never yet seen one? ‘The greatest objection is, that of the Practise; when men ask, where, and when, such Power has by Subjects been acknowledged. But one may ask them again, when, or where has there been a Kingdome long free from Sedition and Civill Warre.’ ²⁴⁹ The implied answer to this rhetorical question is clearly: never and nowhere, as yet. Today's readers of Hobbes tend to overlook that he is presenting a vision and not describing something that was even close to being implemented in his own time, let alone in any earlier period.

Hobbes as pioneer of today's concept of statehood. In previous publications I have been guilty of downplaying Hobbes’ theory of ‘international’, or interstate, relations. I have argued that this alleged theory is limited to scattered bits and pieces, over‐interpreted or indeed misinterpreted by present‐day IR authors. Thus, I have observed that the famous passage in chapter 13 of Leviathan about the natural state of war between ‘kings and persons of sovereign authority’ speaks of relations between individuals, which cannot simply be equated with interstate relations ²⁵⁰ —but it can, as we shall see shortly. It is true that there are only scattered elements of a theory of interstate relations in Hobbes. Yet taken together they do add up to a coherent approach, centred on something that Hobbes at least occasionally calls 'state’ and which is largely equivalent to what is nowadays understood by that term. Moreover, Hobbes’ theory of interstate relations is indeed comparable to the present‐day ‘realist’ view of relations between autonomous actors as, at bottom, necessarily a constant struggle for power between them.

Writing in Latin, Althusius had no word for the 'state’ (paradoxically, since that word is derived from the Latin status). Instead, his generic term for the greatest autonomous political unit is consociatio universalis or universal association, which he uses alongside politia, imperium, regnum, and respublica. In his Politica, they all mean the same thing. Like Bodin, Althusius uses regnum ‘kingship’ and imperium ‘empire’ indiscriminately; Althusius moreover inveighs against the tendency of some authors to employ respublica ‘commonwealth’ as an antonym to ‘monarchy’ (a sense that has since come to prevail: ‘republic’). This refusal to look at whether power in them was shared or not to distinguish types of political association is of course in keeping with Althusius’ notion that power should as far as possible remain with the ‘people’ in any event. ²⁵¹ Bodin, on the other hand, routinely uses the word état (estat) to designate political units, synonymously with république (commonwealth) and royaume (kingdom) (e.g. chapter 2.1: l'estat de France). At the same time, Bodin uses the word to designate the estates within the kingdom (e.g. in the same chapter les trois estats ‘the three estates’, i.e. burgesses, nobility, and clergy). Similarly, Hobbes routinely uses state to mean that which the sovereign is in power over—though again this is not the only meaning of the word in Hobbes.

(p. 443 ) For a long time, 'state’ and ‘estate’ were used interchangeably in English. In Leviathan we still find ‘estate’ where current usage would require 'state’: ‘the estate [=condition] of Man can never be without some incommodity or other,’ or ‘no new evill is thereby on any man Inflicted; he is onely left in the estate he was in before.’ ²⁵² Elsewhere the word evidently means ‘government’ or ‘form of government’, as in the passage already quoted: ‘Greeks and Romanes …living under Popular States’ (where, the context makes clear, Hobbes uses ‘popular’ to render dêmokratikós in Aristotle); likewise (despite the variant spelling) in the passage ‘Monarchy …bitten to the quick, by those Democraticall writers, that continually snarle at that estate.’ Elsewhere again, Hobbes speaks of the pre‐christian supreme priest of the Romans (the pontifex maximus) as ‘an officer subject to the Civill State’. What he means is 'subject to the civil authorities’; ‘civil state’ is used several times in the chapter in question and evidently synonymous with such expressions, also employed in that chapter, as ‘civil dominion’, ‘civil sovereign’, ‘civil power’, and ‘civil laws’. ²⁵³

The Oxford English Dictionary dates the earliest use of 'state’ in the sense of ‘body politic’ to 1538. Hobbes often does employ the word in this sense, treating the 'state’ (body politic) as a bounded community, homogeneous in its subjection to a central power. Bodin clearly separates the sovereign from the 'state’ (état), in the sense of the community of the subjects. For Althusius, the people is sovereign, not the ruler. Hobbes is ambiguous. By virtue of the fictitious treaty of subjection the sovereign is both the creature of those who concluded it (even though, in practice, no one actually concluded it), and their lord—a notion already developed by Nicholas of Kues. At the beginning of his book Hobbes expressly equates both ‘commonwealth’ and 'state’ with Latin civitas ‘citizenry’, as well as with the ‘artificial’ person of the Leviathan or sovereign: ‘that great Leviathan called a Common‐wealth, or State, (in latine Civitas) which is but an Artificiall Man.’ According to Hobbes, ‘every Subject is Author of every act the Soveraign doth.’ ²⁵⁴

This step in the direction of the notion of corporate sovereignty (a notion absent from Bodin, but already present in Althusius) is interesting if also (again) redundant, since in the context of Hobbes’ theory nothing of substance rides on this. Perhaps the idea was to suggest some sort of ‘democratic’ legitimation of the sovereign—which it would be tempting to see as a half‐conscious nod to the monarchomachs, ‘those Democraticall writers, that continually snarle at [Monarchy]’. But elsewhere, as we saw, Hobbes is clear that the sovereign needs no legitimation other than effectiveness. For Hobbes, the sovereign might be either a single person or, less ideally, an assembly; in his text he usually treats the sovereign as a person, more concretely a man (by employing male pronouns).

All power not exercised by the ruler is still his, only delegated: ‘[N]o man but the Soveraign, receiveth his power Dei gratia simply; that is to say, from the favour of none but god: All other [sic], receive theirs from the favour and providence of god, and their Soveraigns.’ ²⁵⁵ The sovereign makes and unmakes the laws without being bound by them:

(p. 444 )

The Legislator in all Common‐wealths, is only the Soveraign. …The Soveraign of a Common‐wealth …is not Subject to the Civill Lawes. For having power to make, and repeale Lawes, he may when he pleaseth, free himselfe from that subjection, by repealing those Lawes that trouble him, and making of new.

For Hobbes, ‘Law, properly is the word of him that by right hath command over others.’ ²⁵⁶ This left no room for customary law:

When long Use obtaineth the authority of a Law, it is not the Length of Time that maketh the Authority, but the Will of the Soveraign signified by his silence. …And therefore …the Length of Time shal bring no prejudice to his Right. ²⁵⁷

Hobbes also disagrees with Bodin that the sovereign cannot alter the law of succession:

There is no perfect forme of Government, where the disposing of the Succession is not in the present Soveraign. ²⁵⁸

All of this was far removed from contemporary reality. No ancien régime monarch could afford to call in question, let alone abrogate in any very aggressive manner, the customary law followed by his subjects—and that meant the vast bulk of the law, since the kind of quasi‐perpetual and all‐encompassing legislative activity typical of today's politics was then unknown. Likewise, however much he might be called a 'sovereign’, a ruler could not easily change the law of succession that had become customary in a given kingdom or principality.

The phrase attributed to the thirteenth‐century English lawyer Henry de Bracton, lex facit regem ‘the law makes the king’, remained valid throughout the ancien régime—an empirical social fact that did not prevent Hobbes from pouring ridicule on the notion of the rule of law (and, once again, on Aristotle while he was at it). Bodin, as we saw, shied away from a radical, consistent approach in this respect: his sovereign was not supposed to touch the law of succession, or the property of his subjects even in the form of taxes, and had to expect passive resistance if his commands were deemed unethical. Hobbes, who evidently prided himself on his rigour, admits no such qualifications. His sovereign really is above the law, legibus solutus:

this is another Errour of Aristotles Politiques [Politiká 1292a: 32], that in a wel ordered Common‐wealth, not Men should govern, but the Laws. What man, that has his naturall Senses, though he can neither write nor read, does not find himself governed by them he fears, and beleeves can kill or hurt him when he obeyeth not? or that beleeves the Law can hurt him; that is, Words, and Paper, without the Hands, and Swords of men? ²⁵⁹

If, in Hobbes’ vision, everybody is not equal before the law, nevertheless they are necessarily equal before the sovereign.

The safety of the People, requireth further, from him, or them that have the Soveraign Power, that Justice be equally administred to all degrees of People; that is, that as well the rich, and mighty, as poor and obscure persons, may be righted of the injuries done them; so as the great, (p. 445 ) may have no greater hope of impunity, when they doe violence, dishonour, or any Injury to the meaner sort, than when one of these, does the like to one of them. …The Inequality of Subjects, proceedeth from the Acts of Soveraign Power; and therefore has no more place in the presence of the Soveraign; that is to say, in a Court of Justice, then the Inequality between Kings, and their Subjects, in the presence of the King of Kings [i.e. god]. …To Equall Justice, appertaineth also the Equall imposition of Taxes. ²⁶⁰

This is strikingly more close to the state of the twentieth century (or its theory, anyway) than to any real existing kingdom of the ancien régime.

Neither Bodin, nor Althusius, nor Hobbes stress any territorial aspect of political community: none of them sees the territorial dimension of politics as important in its own right. On the other hand, political community as conceived by Hobbes is, at least, clearly demarcated: individuals are or are not subjects of a given sovereign, to the same degree and with internal divisions playing no role. One may therefore assume that the community of subjects of a given sovereign is clearly bounded also in a territorial sense—it is not to be expected after all that subjects of the same sovereign will be scattered over non‐contiguous parts of the world, or intermingled with subjects of some other sovereign. Indeed Hobbes speaks of frontiers:

…Kings, and Persons of Soveraigne authority, because of their Independency, are in continuall jealousies, and in the state and posture of Gladiators; having their weapons pointing, and their eyes fixed on one another; that is, their Forts, Garrisons, and Guns upon the Frontiers of their Kingdomes; and continuall Spyes upon their neighbours; which is a posture of War.

And again:

For as amongst masterlesse men, there is perpetuall war, of every man against his neighbour. …So in States, and Common‐wealths not dependent on one another, every Common‐wealth, …has an absolute Libertie, to doe what it shall judge (that is to say, what that Man, or Assemblie that representeth it, shall judge) most conducing to their benefit. But withall, they live in the condition of a perpetuall war, and upon the confines of battel, with their frontiers armed, and canons planted against their neighbours round about. ²⁶¹

Comparison of those two passages shows that Hobbes in discussing the external relations of autonomous political units switches easily between vocabulary designating individuals on the one hand and corporations on the other. Kings and other ‘persons of sovereign authority’ are equated with states and commonwealths. What nowadays seems natural enough was new to the ancien régime; it is of course logical given Hobbes’ (novel) notion that the 'sovereign’ acts for (in place of) his subjects (or they through him). The two could thus be equated, and the state anthropomorphized: ‘A Multitude of men, are made One Person, when they are by one man, or one Person, Represented.’ The reason why the sovereign can be said to represent the subjects is that they have transferred all power they may have held to him (as in ancien régime society they in fact had not). (If the sovereign is not one person but an assembly, then according to Hobbes it is (p. 446 ) the majority within that assembly that on a given issue may be taken to represent the subjects.) ²⁶²

As I said earlier, nothing in Hobbes’ theory actually rides on this equivalence established between the sovereign and the subjects; it is semantic rather than of any substantive importance for the theory. Its ideological potential however was to prove considerable: in our day, the ‘person in the street’ interviewed on the politics of their ‘country’ (meaning, in fact, government) is likely to say ‘we’ when referring to those in power, despite the fact that he or she has no influence on what the government does. (Of course, unlike Hobbes’ sovereign, the government of present‐day states is supposed to defer to the ‘will’ of the people, allegedly expressed by means of elections. But that, too, is a form of words rather than a substantive assertion. At best, elections will have some limited influence on the identity of those in power—limited in that candidates not backed by the political establishment stand no chance—but only in the most indirect and tenuous manner on their decision‐making.) Hobbes effectively treats the sovereign—a person—as if he were the state; but with that equivalence once established it is just as easy to speak of the state as if it were a person. The idea or, perhaps more appropriately, the ideology of the 'state’ as a unitary political actor, which, therefore, may be equated with a person (and, in effect, its government), is a core element of the ideology of the state in the twentieth century and thus, of course, of IR ‘realism’. ²⁶³

On the basis of this conception of autonomous political units, Hobbes sketches a view of interstate relations that seems largely familiar to the early twenty‐first century—because key elements of his thinking suffuse present‐day political culture, which was not the case in his own time. Hobbes thus speaks, as we do nowadays, of (‘foreign’) 'states’ and ‘nations’, of the decision on war and peace belonging to the central power (government), represented by such personnel as ‘ambassadors’ (though to be sure ‘ambassadors’ in the seventeenth century were not quite the same thing as what are so designated now, nor are ‘messengers’ or ‘heralds’ used any longer):

[There] is annexed to the Soveraignty, the right of making Warre, and Peace with other Nations, and Common‐wealths. ²⁶⁴

Publique Ministers abroad, are those that represent the Person of their own Soveraign, to forraign States. Such are Ambassadors, Messengers, Agents, and Heralds, sent by publique Authoritie, and on publique Businesse. ²⁶⁵

A key tenet of Hobbes’ theory of interstate relations is that they are characterized by a perpetual if sometimes latent state of war, with actors forever waiting for opportunities to outmanoeuvre one another in a zero‐sum game. In this game, those without power lead a precarious existence.

And in all places, where men have lived by small Families [which Hobbes regards as the natural form of human community], to robbe and spoyle one another, has been a Trade, and so farre from being reputed against the Law of Nature, that the greater spoyles they gained, the greater was their honour. …And as small Familyes did then; so now do Cities and Kingdomes which (p. 447 ) are but greater Families (for their own security) enlarge their Dominions, upon all pretences of danger, and fear of Invasion, or assistance that may be given to Invaders, endeavour as much as they can, to subdue, or weaken their neighbours, by open force, and secret arts, for want of other Caution, justly; and are remembred for it in after ages with honour. ²⁶⁶

forraign Common‐wealths; who in order to the good of their own Subjects let slip few occasions to weaken the estate [= position? government? state?] of their Neighbours. ²⁶⁷

And as for very little Common‐wealths, …there is no humane wisdome can uphold them, longer then the Jealousy lasteth of their potent Neighbours. ²⁶⁸

The last sentence seems to imply that units incapable of defending themselves only remain in existence because of a stand‐off, a balance of power, between stronger units that begrudge each other the expansion resulting from absorption of weaker units. Equilibrist thinking is also found elsewhere in the book:

The Multitude sufficient to confide in for our Security, is not determined by any certain number, but by comparison with the Enemy we feare; and is then sufficient, when the odds of the Enemy is not of so visible and conspicuous moment, to determine the event of warre, as to move him to attempt. ²⁶⁹

Since, for Hobbes, law can only either be natural law or consist of the enforceable decrees of the sovereign, there can of course be no ‘international law’. ‘Where there is no common Power, there is no Law: where no Law, no Injustice,’ ²⁷⁰ Hobbes remarks with regard to individuals—but by the logic of his theory the phrase is fully applicable to 'states’ as well: they are in the pre‐social state of nature in which only natural law applies. This, for Hobbes, means that anything is allowed in order to ensure one's survival. In the traditional view of the period, natural law comprised rules of coexistence and of the treatment of others to be found among all humans (hence the traditional identification of natural law with the ‘law of [all] nations’, ius gentium). Hobbes rejects that tradition and redefines the concept completely. He retains the notion that natural law is what reason dictates, but, for him, this is not respect for others (or their lives, property, etc.), but only respect for oneself.

A LawOf Nature, (Lex Naturalis,) is a Precept, or generall Rule, found out by Reason, by which a man is forbidden to do, that, which is destructive of his life, or taketh away the means of preserving the same; and to omit, that, by which he thinketh it may be best preserved. …The Right Of Nature, which Writers commonly call Jus Naturale, is the Liberty each man hath, to use his own power, as he will himselfe, for the preservation of his own Nature; that is to say, of his own Life; and consequently, of doing any thing, which in his own Judgment, and Reason, hee shall conceive to be the aptest means thereunto. ²⁷¹

This natural law, in the sense of a natural right to defend one's own existence without regard for others, ends as soon as there is a sovereign to take over the protection of individuals that hitherto they had to ensure themselves. Between sovereigns (states) (p. 448 ) it subsists. In this context Hobbes explicitly draws a parallel between individuals and sovereigns/commonwealths.

Concerning the Offices of one Soveraign to another, which are comprehended in that Law, which is commonly called the Law of Nations [=ius gentium], I need not say any thing in this place; because the Law of Nations, and the Law of Nature, is the same thing. And every Soveraign hath the same Right, in procuring the safety of his People, that any particular man can have, in procuring the safety of his own Body. And the same Law, that dictateth to men that have no Civil Government, what they ought to do, and what to avoyd in regard of one another, dictateth the same to Common‐wealths, that is to the Consciences of Soveraign Princes, and Soveraign Assemblies. ²⁷²

…every Common‐wealth, …has an absolute Libertie, to doe what it shall judge (that is to say, what that Man, or Assemblie that representeth it, shall judge) most conducing to their benefit. ²⁷³

Hobbes’ image of interstate relations: not analysis, but ideology. Hobbes’ depiction of interstate politics is derived not from empirical observation of Hobbes’ own present or of the past (there is no discussion of historical events in Leviathan), but from a desire to justify a radical normative political theory, an ideology. Hobbes needs the state of war between individuals that he posits to legitimize his omnipotent sovereign. It is only because individuals are irredeemably evil, and incapable of constructive coexistence, that depriving them of all freedom can be defended at all.

Inevitably, such a radical stance must provoke doubts. Are men really so bad? Hobbes counters such doubts by adducing various observations from everyday life (Leviathan, chapter 13). The comparison between pre‐social ordinary individuals and ‘kings, and persons of sovereign authority’ is the capstone of this reasoning. If readers still hesitated to accept that a state of nature as described by Hobbes ever actually existed, all they had to do was look at the relations between sovereigns, who for the purpose of the argument had the merit of still existing when other pre‐social individuals at least in Europe did not. In fact, Hobbes himself was so little troubled by the prospect of foreign war that he saw no need for ending the state of nature as between kings by creating a global Leviathan.

Kings, and Persons of Soveraigne authority …are …in the state and posture of Gladiators; …which is a posture of War. But because they uphold thereby, the Industry of their Subjects; there does not follow from it, that misery, which accompanies the Liberty of particular men. ²⁷⁴

Hobbes even thought that the state of war might be a good thing to prevent domestic strife: ‘when there is no common enemy, they [the multitude] make warre upon each other, for their particular interests’; and again: ‘no great Popular [i.e. democratic] Commonwealth was ever kept up; but [e.g.] by a forraign Enemy that united them.’ ²⁷⁵ One may wonder, of course, how it can even be that a commonwealth lacks a ‘foreign enemy’, if it is true that it is automatically in a state of war with its neighbours.

(p. 449 ) Hobbes’ image of interstate relations is no more based on an analysis of real social phenomena than his depiction of domestic society as characterized by universal unconditional subjection to the prince, or his contention that humans could only be in one of two situations: war of all against all or complete subjection. We have seen that, in fact, western civilization has known stages dominated by forms of cooperative social organization rather than rule, such as notably the Greek pólis; nor were the towns that played such a crucial role in the evolution of Latin christendom after the turn of the millennium based on subjection in their internal organization. In rejecting the notion that laws could be effective in the absence of a coercive central power Hobbes is historically wrong: as noted there was no such power for example in Greek society, which nevertheless found laws useful. Closer in time to Hobbes, Althusius paints a more realistic picture of seventeenth‐century society as a complex conglomerate of nested communities each of which retained considerable autonomy—on which more in Section 4.4 of this chapter.

Concerning the domestic situation, Hobbes himself concedes explicitly that his vision is utopian, an ideal that would not become reality until people had grasped Hobbes’ theory. Hobbes hoped that this would happen:

considering how different this [Hobbes’] Doctrine is, from the Practise of the greatest part of the world, especially of these Western parts, that have received their Morall learning from Rome, and Athens; and how much depth of Morall Philosophy is required, in them that have the Administration of the Soveraign Power; I am at the point of believing this my labour, as uselesse, as the Common‐wealth [‘Republic’] of Plato; For he also is of opinion that it is impossible for the disorders of State, and change of Governments by Civill Warre, ever to be taken away, till Soveraigns be Philosophers. But when I consider again, that …neither Plato, nor any other Philosopher hitherto, hath put into order, and sufficiently, or probably proved all the Theoremes of Morall doctrine, that men may learn thereby, both how to govern, and how to obey; I recover some hope. ²⁷⁶

By contrast, Hobbes presents his view of interstate relations as grounded in reality: the alleged state of war, and thus the state of nature, among kings is presented as empirical evidence of what would be the situation of all individuals were there no kings. However, the contention that there is indeed a permanent state of war between kings only works by virtue of some semantic legerdemain: ‘the nature of War, consisteth not in actuall fighting; but in the known disposition thereto, during all the time there is no assurance to the contrary.’ ²⁷⁷ This is not a very useful notion (useful, that is, other than for Hobbes’ advocacy of a strong sovereign): no matter how long an absence of violent conflict lasts, or between how many sovereigns, by this definition it is still war since no one can guarantee its further continuance. Even a pledge by sovereigns to abstain from war, or indeed an alliance between them, could not end the state of war between them, since for Hobbes a treaty that cannot be enforced has no value: ‘Covenants, without the Sword, are but Words, and of no strength to secure a man at all.’ As with the state of war, this kind of assertion may be said to be corroborated in logic if one treaty out of a hundred is broken; and yet this defies common sense. ²⁷⁸