From Caesar to Tacitus: changes in early Germanic governance circa 50 BC-50 AD

Abstract

Julius Caesar and Cornelius Tacitus provide characterizations of early Germanic (barbarian) society around, respectively, 50 BC and 50 AD. The earlier date corresponds to expansion of Rome to the Rhine and Danube. During the subsequent century Germanic governance institutions changed in a number of ways. In particular, (1) temporary military commanders elected from the nobility gave way to standing retinues under the leadership of professional commanders, (2) public assemblies met more frequently and regularly, (3) councils made up of nobility gained agenda control in the assemblies, and (4) these councils relinquished their control over the allocations of land. I account for these constitutional exchanges in light of Rome’s encroachment. This encroachment brought new sources of wealth as well as constraints on the expansion of Germans into new lands. Incentives favored a reallocation of resources away from pastoralism and towards both sedentary farming and raids across the frontier.

1 Introduction

The emergence of governance from anarchy has long intrigued political thinkers.¹ Hobbes, Locke, and Rousseau each employed state of nature thought experiments to ask how a just government might arise from initially anarchic conditions. Political philosophers such as Nozick (1974) and Rawls (1999 [1971]) followed their lead, as did public choice economists who emphasized social contracts (Buchanan and Tullock 1962; Buchanan 1975).²

The authors referenced above all focused on the emergence of government.³ However, government is only one source of governance. Numerous scholars have documented the emergence of constitutional frameworks for self-governance in environments where effective government authority is lacking. These environments include prisons (Skarbek 2010, 2011, 2014), the nineteenth century American West (Anderson and Hill 2004), terrorist organizations (Berman 2009; Berman and Laitin 2008), seafaring piratical communities (Leeson 2007b, 2009a), medieval maritime activities (Grief 1993; Fink 2011) and financial markets (Stringham 2002, 2003).⁴

These studies are important contributions. However, they are based on pockets of anarchy embedded within or adjacent to actual governments.⁵ The cowboys, prisoners, pirates, terrorists, and traders they describe lived under the shadows of formal governments. Their experiences inevitably informed and shaped their institutions of self-governance. Ideally we could observe the emergence and evolution of self-governance among individuals lacking government-based templates.

Unfortunately, true and observable states of nature are difficult to find.⁶ The most interesting cases are prehistoric. In this paper I exploit an instance of an historic society coming into contact with a prehistoric one. In particular, I take advantage of two remarkable accounts of Germanic barbarians by Julius Caesar and Cornelius Tacitus. These accounts bookend a period of roughly 100 years (50 BC to 50 AD) that begins with Caesar’s conquest of Gaul and the establishment of the Rhine and Danube rivers as Roman frontiers.⁷ Beyond these rivers lay Germania, a vast area that encompassed modern Germany, Denmark, Poland, Slovakia, the Czech Republic, half of Hungary, and a part of Austria. (See Fig. 1).⁸

Fig. 1

Caesar’s campaigns in Gaul (left) and the Roman Empire and Germania circa 100 AD (right). Notes Caesar’s campaign against Ariovistus in 58 BC is indicated by the two arrows farthest to the southeast. Both maps are in the public domain and were last accessed on February 7, 2015: http://commons.wikimedia.org/wiki/File:Caesar%27s_Campaigns_in_Gaul,_1st_century_BC.png; http://commons.wikimedia.org/wiki/File:Imperium_Romanum_Germania.png

Germanic barbarians did not have governments but did have institutions of self-governance. From Caesar to Tacitus we get a glimpse of those institutions and how they changed in the intervening century. Caesar describes tribes and confederacies that had little in the way of constitutions. The barbarians relied on conventions that facilitated infrequent political action of a limited range. Alternatively, Tacitus’ account describes governance institutions that had evolved in important ways to accommodate political action of both greater range and frequency.

Caesar’s conquest of Gaul changed economic conditions in ways that were fundamental to the observed changes in Germanic governance. However, those changes primarily were not means to avoid subjugation by the Romans. Rome’s halt at the Rhine often is incorrectly attributed to the Battle of the Teutoburg Forest (9 AD). During that battle the Cheruscian chieftain Arminius and his followers routed three Roman legions (20,000 to 30,000 men, including auxiliary troops).⁹ Notwithstanding Arminius’ David-versus-Goliath-like victory, the Roman general Germanicus exacted quick revenge in 16 AD.¹⁰ In truth, “Arminius had won one huge, fluke victory, but the underlying reasons for the halting of the legions on the fringes of first-century Germania were altogether different” (Heather 2006, p. 55).

In particular, the Rhine corresponded to an intermediate zone between what archaeologists refer to as the La Tène and Jastorf material cultures. Larger, denser settlements and greater economic development characterized the La Tène culture:

[C]oins were in use, and some of its populations were literate. […] All of this rested on an economy that could produce sufficient food surpluses to support warrior, priestly and artisan classes not engaged in primary agricultural production. Jastorf Europe, by contrast, operated at a much starker level of subsistence, with a greater emphasis on pastoral agriculture and much less of a food surplus. Its population had no coinage or literacy, and […] no substantial settlements. (Heather 2006, pp. 56–57).

Roman expansion halted “around a major fault-line in European socio-economic organization” beyond which “the difficulties involved in incorporating the next patch of territory, combined with the relative lack of wealth that [could] be extracted from it, [made] further conquest unattractive” (Heather 2006, p. 57).¹¹

While Roman encroachment did not credibly threaten the barbarians with subjugation, it did expose them to new sources of wealth and also constrained their expansion into new lands. A century later, Tacitus describes a number of changes in Germanic governance institutions: (1) temporary military leaders elected from the nobility yielded to standing retinues led by professional commanders; (2) public assemblies met more frequently and regularly; (3) councils comprised of the nobility had acquired agenda control in the public assemblies; and (4) the nobility had relinquished control over land allocations. According to Congleton (2011a, p. 9): “Mutual gains from constitutional exchange occasionally emerge, which can be realized by amending the preexisting constitutions.” I argue that the changes in Germanic institutions are made intelligible by constitutional exchanges that benefited both the nobility and freemen. These exchanges occurred against a backdrop of incentives that favored a reallocation of resources away from pastoralism and towards both sedentary farming and raiding activities across the frontier.¹²

Historians have noted the institutional differences across the accounts of Caesar and Tacitus (e.g., Thompson 1965). However, to my knowledge I provide the first economic analysis of non-Roman, pre-medieval political institutional change in Europe. In accounting for the institutions described in the Commentarii de Bello Gallico versus those described in the Germania, I focus on decision-making rules and divisions of governance (checks and balances). These rules and divisions provided foundations for the early medieval successor kingdoms.

Hadfield and Weingast (2013) and Kerekes and Williamson (2012) both provide analyses of the legal institutions of medieval Iceland. Those papers are related to mine in that they explore the governance institutions of (what Romans would have considered to be) northern barbarians. They describe the functioning of those institutions in the absence of a centralized, coercive state. However, Iceland was settled about 900 years after the period studied here. Also, the Icelandic legal institutions remained largely unchanged from the 10th through the Thirteenth centuries. I study how Germanic self-governance institutions evolved in response to Rome’s encroachment.¹³

This paper also is related to the work of Fleck and Hanssen (2006). Focusing on ancient Greece, they argue that exogenous and costly-to-monitor variation in the relative returns to agricultural investments in Athens versus Sparta can help to account for the former being more democratic. Fleck and Hanssen focus on the severity of time-consistency problems and whether or not the elites will find it worthwhile to make a commitment by extending political rights to the non-elite (a la Acemoglu 2003; Acemoglu and Johnson 2000, 2001). I too examine constitutional exchanges between Germanic elites and non-elites but with tighter focus on incremental exchanges within an already divided set of governance institutions (Congelton 2007, 2011a).

In Sect. 2, I elaborate on the Germanic societies observed by Caesar circa 50 BC. Then I do the same in Sect. 3 for the circa 50 AD societies described by Tacitus. I summarize the salient changes in Germanic self-governance from Caesar to Tacitus in Sect. 4. My analysis of these changes in terms of constitutional exchange occurs in Sect. 5. I conclude in Sect. 6.

2 Germanic peoples circa 50 BC

Julius Caesar published his Commentarii de Bello Gallico (Gallic War) following his military campaigns from 58 BC to 50 BC. These campaigns led to the establishment of the Rhine and the Danube as Rome’s approximate frontiers. Following his defeat of the Helvetti, Caesar came into conflict with the Suebi, the “largest and most warlike nation among the Germans”. Caesar recorded numerous observations of these Germans and their society.

A king (rex) named Ariovistus led the Suebi (Caesar book IV, p. 181; book I, pp. 49–87). They were not a specific tribe, but rather a confederacy of tribes—or individuals from different tribes—united towards specific ends. The Suebian confederacy had seized and settled land across the Rhine in Gaul and was threatening further grabs:

Ariovistus, king of the Germans, has settled within [the Sequani’s—a Gallic group] borders and seized a third of the part of their territory, the best in all Gaul; and now he orders them to evacuate another third, because a few months since 24,000 of the Harudes [a Germanic tribe] joined him, for whom he had to provide settlement and a home. (Caesar book I, p. 49)

Caesar attributes to the Suebi “a hundred cantons from which they draw one thousand armed men yearly” and the settlement of 120,000 Germans across the Rhine in Gaul (Caesar book IV, p. 181; book I, p. 47).¹⁴ Todd (1987, p. 5) remarks: “It is probable that the total population of Germania at the beginning of the Roman Iron Age was more than one million, but less than two.” The Suebi, then, were drawn from a number of different tribes and constituted a non-negligible part of the total barbarian population.

Furthermore, Tacitus also speaks of the Suebi in his Germania. A 100 years later they still appear to be a confederacy drawing from numerous Germanic tribes:

We must now speak of the Suebi, who do not, like the Chatti or the Tencteri, constitute a single nation. They occupy more than half of Germany, and are divided into a number of separate tribes under different names, though they all are called by the generic title of “Suebi”. (Tacitus Chap. 38, p. 133)

This passage supports the idea that Caesar’s observations generalize across Germanic barbarian groups; also that Tacitus was commenting later on groups that were descendant from those observed by Caesar.

Caesar’s recordings make it clear that Ariovistus was not a standing king. Rather, he was a member of the nobility who had been granted military command temporarily:

When a state makes or resists aggressive war officers [magistratus] are chosen to direct the same, with the power of life or death. In time of peace there is no general officer of state [communis magistratus], but the chiefs [princepes] of the districts and cantons do justice among their followers and settle disputes. Acts of brigandage committed outside the borders of each several state involve no disgrace[….] And when any of the chiefs has said in public assembly [concilio dixit] that he will be leader [dux], “Let those who will follow declare it,” then all who approve the cause and the man rise together to his service and promise their own assistance[.] (Caesar book VI, p. 349)

A king such as Ariovistus was a temporary leader chosen from the nobility (or leading men) and, having put forth his cause at a public assembly, elected by voluntary consent of the free population. Such elections appear only to have occurred during times of war or to organize aggressive, wealth-extracting raids on foreigners.¹⁵ Only in the case of war did they actually have coercive “power over life or death”. Otherwise such kings relied on the oaths of their followers.

The “public assembly”, at which a princeps could seek election to the role of duc by acclamation, is the only mention of a general gathering of freemen; something akin to the Assembly of Warriors later described by Tacitus. (See Sect. 3 below). Otherwise we find reference in Caesar’s account only to a more exclusive council of the leading men from various tribes. Such a council appears largely to be analogous to the Council of Leading Men that we encounter in Tacitus’ account. (Again, see Sect. 3 below).

Regarding the council, Caesar (book VI, p. 347) describes only one regular meeting:

No man has a definite quantity of land or estate of his own: the magistrates and chiefs [magistratus ac principes] every year assign to tribes and clans [gentibus congnationibusque hominum] that have assembled together as much land and in such place as seems good to them, and compel the tenants after a year to pass on elsewhere.

Aside from an annual meeting to allocate land, Caesar (book IV, p. 205) suggests that council meetings occurred only under extraordinary circumstances, such as during a military threat:

The Suebi, when they had discovered by means of their scouts that a bridge was being built [by Caesar to cross the Rhine], held a convention [concilio] according to their custom, and dispatched messengers to all quarters, ordering the people […] to assemble in one place all men capable of bearing arms.

This governance body thus played a regular role in facilitating intertribal collective action only once a year and otherwise only during emergencies: “In peacetime no council higher than the councils of the pagi [individual tribes or villages] can be said with certainty to have existed” (Thompson 1965, p. 13).¹⁶

Overall, Caesar paints a picture of a Germanic society with governance institutions that, beyond the individual tribe or clan, facilitated infrequent political action of a very limited range.

The kindred [tribe or clan] […] was the fundamental entity in society. The kindreds only loosely knit together, and in peacetime there seems to have been no public authority to weld them all into a unity. […] [T]here were no public institutions of coercion. In all, so primitive was Germanic society that we cannot disregard Caesar’s remark that the life of the Germans was one of poverty, want, and hardship. (Thompson 1965, p. 17)

Legitimized intertribal coercion was absent. The Germanic barbarians did not have governments. Furthermore, circa 50 BC the institutions of self-governance were minimal. Relative to what researchers typically can observe Germania was not far removed from a state of nature.

3 Germanic peoples circa 50 AD

Gaius Cornelius Tacitus published his Germania around 98 AD.¹⁷ However, it is likely that he never set foot in Germania and relied heavily on the Elder Pliny’s lost Bella Germaniae (Gudeman 1900). This suggests that Tacitus’ characterization of Germanic societies applies to the period between 50 and 54 AD. Of course, we cannot rule out the influence of other sources and, perhaps, even Tacitus’ own experiences, so some details contained with the Germania may be based on the later first century AD. Still, the Germania provides an account of Germanic governance 100–150 years after Caesar’s observations.

Tacitus provides a wealth of details regarding the political institutions of the Germans. They fit into the broad contours of the King and Council template of divided government elaborated on by Congleton (2001, 2011a)

neither an unrestricted executive (leviathan or dictator) nor an unrestrained parliament (legislature, council, committee, or diet). […] It divides up policy-making responsibility between a branch of government headed by one person, and another branch in the form of a committee composed of several members having more or less equal authority. I refer to this very general and ancient constitutional template as “king and council”. (Congleton 2001, p. 193)

I mention the broad contours of this template because the barbarians had both general councils that included all freemen and more exclusive executive councils. Using the language of Thompson (1965), I shall refer to the general and executive councils, respectively, as the Assembly of Warriors and Council of Leading Men. In addition to the Assembly and Council, the barbarians of Tacitus’ Germania had two types of kings: one elected based on merit (dux) and the other based on noble birth (rex). Tacitus draws a clear distinction between these two types of kings, the former of which was “a later, more developed, and less egalitarian form of institution” that has “no parallel to it in Caesar” (Thompson 1965, p. 33).

3.1 Assembly of warriors

The Assembly of Warriors was the more popular of the governance structures. It included all male citizens, both freemen and nobles (leading men). Germanic society also comprised slaves and freedmen who were not counted as citizens. Slaves and freedmen factor directly into none of the three governance entities considered here (Assembly, Council, and Kings). (I return to consider their role in Germanic society in Sect. 3.4 below).¹⁸

I follow Thompson (1965) in appending “of Warriors” to what Tacitus simply refers to as the assembly (concilium). The appendage is descriptive. Bearing of arms and military obligation were intrinsic to membership in the Assembly of Warriors. The rite of passage into manhood involved the bestowing of weapons “in the presence of the Assembly”: “the first distinction publicly conferred upon a youth, who now ceases to rank merely as a member of a household and becomes a citizen” (Tacitus Chap. 13, p. 112). Once weapons were bestowed, to “throw away one’s shield in battle is the supreme disgrace, and the man who has thus dishonoured himself is debarred from attendance at sacrifice or assembly” (Tacitus Chap. 6, pp. 106–107).

The Assembly of Warriors met approximately monthly: “on certain particular days either shortly after the new moon or shortly before the new moon” (Tacitus Chap. 11, p. 110). The Assembly debated “major affairs” (as opposed to “matters of minor importance” that could be taken up independently by the Council of Leading Men—see below). These likely included declarations of peace or war, as Caesar described earlier (Thompson 1965, p. 31; Caesar book VI, p. 205). The Assembly also was “competent to hear criminal charges, especially those involving the risk of capital punishment” and could “elect, among other officials, the magistrates [principes] who administer justice in the districts and villages [pagos vicosque]” (Tacitus Chap. 12, pp. 111–112). Decisions were made by acclamation: “If a proposal displeases them, the people shout their dissent; if they approve, they clash their spears” (Tacitus Chap. 11, p. 111). Criminal charges could result in fines, some of which would go towards restitution for victims; some would be allocated to the nobility. (This last point will be elaborated on in Sect. 5 below).

3.2 Council of leading men

While the Assembly of Warriors had say on major affairs, a Council of Leading Men independently could handle “matters of minor importance” (Tacitus Chap. 11, p. 110). Furthermore, “even where the commons [i.e., the Assembly of Warriors] have the decision, the subject is considered in advance by the chiefs [principes]” (Tacitus Chap. 11, pp. 110). As such, while the Assembly of Warriors ultimately decided major affairs, the Council of Leading Men controlled the agenda of affairs taken up by the Assembly. Tacitus (Chap. 11, p. 111) also describes a chief among chiefs who served as an executive or spokesman for the Council:

When the assembled crowd thinks fit, they take their seats fully armed. […] Then such hearing is given to the king or state-chief [rex vel princeps] […]—more because his advice carries weight than because he has the power to command.

Through this rex vel princeps, the Council of Leading Men set an agenda for the Assembly of Warriors. Given that agenda, the Assembly rendered decisions on its items.

3.3 Kings: reges and duces

Tacitus (Chap. 7, p. 107) records that the Germanic peoples “choose their kings [reges] for their noble birth, their commanders [duces] for their valour”:

The power even of the kings [regibus] is not absolute or arbitrary. The commanders [duces] rely on example rather than on the authority of their rank—on admiration they win by showing conspicuous energy and courage and by pressing forward in front of their own troops. Capital punishment, imprisonment, even flogging, are allowed to none but the priests.

Unlike in Caesar’s account, Tacitus (Chap. 10, p. 110) mentions priests specifically on several occasions. Tacitus associates both the priests and the reges with the sacral:

[The Germanic peoples] try to obtain omens and warnings from horses. These horses are kept at the public expense in the sacred woods and groves [….] The priest and the king, or the chief of state [rex vel princeps civitatis comitantur], yoke them to a sacred chariot and walk beside them taking note of their neighs and snorts.¹⁹

As mentioned above, at the Council of Leading Men such a king spoke with authority “more because his advice carries weight than because he has the power to command (Tacitus Chap. 11, p. 111).

Alternatively, duces are never associated with the sacral in Tacitus’ account. They are instead portrayed as violent formeteurs: individuals who organized warriors into retinues that acted as profit-seeking entities (Congleton 2011b). A commander organized a retinue of warriors as a means of “war and plunder” (Tacitus Chap. 14, p. 113). In other words, these retinues provided defense services and/or pursued wealth extraction.

A German is not so easily prevailed upon to plough the land and wait patiently for harvest as to challenge a foe and earn wounds for his reward. He thinks it tame and spiritless to accumulate slowly by the sweat of his brow what can be got quickly by the loss of a little blood. (Tacitus Chap. 14, pp. 113–114)

Tacitus may exaggerate the Germans’ disdain for productive activities, but we can conclude that the returns to violent rent extraction often exceeded those associated with the peaceful creation of wealth. These returns could be from raids, which were purely extractive, or from the provision of defense services to other Germans facing a threat.

The oath-bound relationship between a commander and his retinue clearly differed from the kin-based relationship between a king and his clan. A Germanic rex had a measure of authority through his noble birth and association with the sacral. For Romans like Tacitus the word rex “implied a moral content: a king should be able to rule himself as well as others” (Wallace-Hadrill 1971, p. 3). Alternatively, a warrior swore obedience to his commander only because he stood to gain by doing so. Tacitus (Chap. 13, p. 112) describes a competitive labor market wherein the supply of warriors interacted with the demand for their services: “there is great rivalry, both among the followers to obtain the highest place in their leader’s estimation and among the chiefs for the honour of having the biggest and most valiant retinue”. Tacitus (Chap. 14, p. 113) refers to the place in the leader’s estimation and honour but his following remarks suggest more mundane motives:

[A] large body of retainers cannot be kept together except by means of violence and war. They are always making demands on the generosity of their chief […]. Their meals […] count in lieu of pay. The wherewithal for this openhandedness comes from war and plunder.

In an effectively organized retinue, a warrior’s marginal product could be larger than that associated with non-violent pursuits. If a particular dux failed to raise a warrior’s marginal product and/or remunerate him consistent with it, offering to serve a competing duce was an available option.

The descriptions above suggest that these retinues functioned as clubs with self-enforcing constitutions (Leeson 2011).²⁰ The commander organized a group of warriors and provided them with leadership in identifying and pursuing profit opportunities. Operating within a competitive market, warriors remained in a particular retinue only so long as it was in their interest to do so. They were residual claimants on that retinue’s revenues (distributed under the guise of the commander’s largesse).

A retinue’s revenues depended on each warrior following the commands of his dux and not shirking during battle. One can imagine internal mechanisms by which a warrior’s shirking would have been punished (likely involving physical violence by other warriors). Furthermore, recall Tacitus’ (Chap. 6, pp. 106–107) remark that for a warrior to “throw away one’s shield in battle is the supreme disgrace, and the man who has thus dishonoured himself is debarred from attendance at sacrifice or assembly”. External mechanisms, then, may also have contributed to enforcing retinues’ constitutions. Shirking may have meant exclusion from religious practices as well as the extra-retinue institutions of governance.²¹

While never mentioned by Tacitus explicitly, it would stand to reason that the “supreme disgrace” also entailed being excluded from other retinues. In that case, a multilateral reputation mechanism would have imposed additional discipline (Grief 1993; Clay 1997; Stringham 2003). The cost incurred by a warrior for shirking would have included not only the exclusion from his current retinue, but also future exclusion from other retinues. The Assembly of Warriors may have served as a forum for information sharing by retinue commanders about shirkers.

The rise of Germanic duces turned out to be a significant development in European history. These commanders were the forerunners of barbarian kingships that supplanted the Western Roman Empire in the fifth and sixth centuries. As Wolfram (1997, p. 15) states: “Though [Tacitus] appears to describe the simultaneous existence of royal and military (ducal) authority, he was in fact recording two forms of Germanic kingship that supplemented, indeed succeeded, one another.” Thompson (1965, p. 50) adds that by the end of the first century a “leader of a retinue had […] transformed his relationship with his companions (comites) into something like a permanent one.” And according to Heather (1998, p. 66), “the rise of groups of specialist armed retainers was a social development of the greatest importance.” Archaeological evidence suggests that these retinues increasingly became specialized and hierarchical over the following three centuries (Heather 1998, pp. 66–68). By the fourth century, confederacies of barbarian retinues were able to extract imperial subsidies and, in some cases, demand settlement and legal autonomy within the Empire’s frontiers.²²

3.4 Slaves and freedmen

After summarizing the changes in Germanic governance institutions that occurred from approximately 50 BC to 50 AD in Sect. 4 below, I will then provide an account of the constitutional exchanges that facilitated those developments in Sect. 5. That account will focus on the governance institutions described above, along with the competing interests of the freemen and nobility. However, as mentioned at the beginning of Sect. 3.1, Germanic society also included slaves and freedmen. What were their interests? What roles did they play in Germanic society? How large a part of that society did they constitute?

Thompson (1957) argues that slaves and freedmen likely represented only small fractions of the early Germanic societies that Caesar and Tacitus described. He builds his argument on the following facts: (a) Tacitus (Chap. 15, p, 114) notes the absence of household slavery, “the care of the house, home, and fields being left to women, old men, and weaklings of the family”; (b) the difficulty of preventing slaves from escaping; and (c) the tendency of slaves to be conveyed towards the frontier for sale to Romans. Caesar (book VI, p. 183) also observed that the Suebi “give access to traders […] to secure purchasers for what they have captured in war[.]” Prisoners of war could profitably be sold across the Rhine. Alternatively, employing slaves’ labor services in Germania was costly because it was difficult to prevent their escape:

If [the slaves] were prisoners of war their kinsmen might live not very far away and so would not be altogether inaccessible. Since there were vast areas of uncultivated land and of forest and marsh a fugitive could very often elude his pursuers and make his escape to his own people. (Thompson 1957, p. 22)

Thompson—aided by what little information Caesar and Tacitus provide—paints a picture of Germanic society wherein most slaves were prisoners of war, destined for sale to individuals beyond the frontier. Freedmen would have been the legacy of the small number of slaves that remained in Germania and they likely would have been owned by a few wealthy leading men.

Perhaps the strongest support for the idea that slaves and freedmen were relatively few in number comes from the silence of both Caesar and Tacitus regarding their role in society. Given the observations that both authors make on the roles of freemen and nobility in governance, why would they ignore completely a considerable percentage of the population? Even given the de jure subservient status of slaves and freedmen, if they had been of substantial numbers they would have wielded considerable de facto political power (Acemoglu and Robinson 2008). They presumably also would have been either a meaningful chunk of the barbarian military forces or a sizeable source of manpower to be turned on their Germanic masters. It is difficult to believe that this would not have been of interest to Caesar in particular.

4 From Caesar to Tacitus: summarizing the changes

The changes in early Germanic governance institutions that occurred from approximately 50 BC to 50 AD are summarized in Table 1. The popular Assembly of Warriors began to meet more frequently and regularly. Also, a more exclusive and elite Council of Leading men gained autonomous authority over a limited range of judicial matters, as well as agenda-setting power in the larger Assembly.

Table 1 Germanic governance institutions around 50 BC and around 50 AD

A shift in executive power from tribal-based, sacral kingships and towards retinues organized under military commanders was underway. During Caesar’s time, military commanders were drawn from royal (or at least noble) families only at times of emergency. Their tenures were both short-lived and irregular. However, Tacitus describes commanders with standing retinues that provided protection services during war and pursued profit-seeking raids as a matter of course. These retinues were constitutionally self-enforcing and their commanders’ tenures were founded on warriors’ respect for their organizational skills and war-making prowess.

In broad strokes, what we witness from Caesar to Tacitus is twofold. First, we observe the regularization and specialization of legislative and judicial governance in (approximately) monthly meetings of the Assembly of Warriors, with minor matters and agenda-setting power delegated to the more exclusive Council of Leading Men. Second, we observe the secularization and constitutionalization of executive power. Whereas a rex wielded authority based on his noble lineage and association with the sacral (and in actual practice exercised executive power only during times of war), duces organized standing retinues. Obedience to a commander was constitutionally self-enforcing based on the arrangement being mutually beneficial to commander and warriors alike.

Regarding the powers of the Assembly of Warriors vis-à-vis the Council of Leading Men, the former gained the rights to adjudicate major criminal charges and decide on “major” affairs. Though not elaborated on by Tacitus, these major affairs likely included the allocation of land, previously a power delegated to the Council. Whether annual reallocations to the villages remained in effect, as Caesar described, is unclear. Tacitus (Chap. 26, p. 122) states:

Lands proportioned to their own number are appropriated in turn for tillage by the whole body of tillers. They then divide them among themselves according to rank; the division is made easy by the wide tracks of cultivable ground available. These ploughlands are changed [mutant] yearly, and still there is enough to spare.

Here, “the whole body of tillers” suggests a decision made by the broad assembly. The word mutant may mean exchanged, altered, or moved. Given that Tacitus states that “there is enough [land] to spare”, he may be referring simply to a relatively loose pattern of crop rotation. Tacitus makes no explicit reference to tenants being “compelled to move on”, as had been the case with the Council of Leading Men as described by Caesar.

Tacitus (but not Caesar) also mentions that the Assembly of Warriors appointed the village magistrates from amongst the leading men. Alternatively, while the leading men could merely call the Assembly to meet during the time of Caesar, a century later the Council exercised agenda control at the Assembly’s regular meetings.

While the Council gained agenda control in the Assembly, its membership of principes lost more than the power to make annual allocations of arable land. The emergence of commanders elected on the basis of military and organizational skills was a signal loss to the nobility. In Caesar’s time, the commanders of raids or wartime expeditions were elected exclusively from the principes if not the reges. Tacitus, however, describes competition amongst profit-seeking retinues wherein success depended on the specialized formeteur/entrepreneur dux.

The transition from kings (noble and sacral) to commanders (formeteur/entrepreneur) led to the development of sophisticated, hierarchically structured and constitutionalized rent-seeking organizations that are reminiscent of the “roving bandits” described by Olson (1993). From the perspective of the communities they raided (e.g., in the Gallic provinces west of the Rhine) the comitati certainly were bandits. Yet they themselves emerged non-coercively from the institutions of early Germanic self-governance. In Caesar’s time, a leading man could aspire at the public assembly to command a raid: “Let those who will follow declare it.” A century later duces led standing, profit-seeking retinues.²³

5 Accounting for the changes in early Germanic governance

From Caesar to Tacitus, the principal institutional changes to account for are the:

• transition from temporary military commanders elected from amongst the nobility to standing retinues under the leadership of formeteur/entrepreneurs;

• more frequent and regular meetings of a public assembly;

• granting of agenda control in the public assembly to the nobility;

• relinquishing of land allocation decisions by the nobility.

I argue that Germanic barbarians were exposed to new sources of wealth and faced greater scarcity of land following Caesar’s conquest of Gaul. The encroachment of Rome created incentives to allocate resources away from pastoralism and towards (a) sedentary agriculture and (b) raiding across the frontier. Constitutional exchanges by which the nobility ceded its control of the land and its claim to military leadership while freemen granted the nobility regular agenda control over the public assembly were mutually beneficial.

The Assembly of Warriors included all freemen; the Council of Leading Men included only the nobility. In Caesar’s time, both kings and commanders were nobles. By 50 AD nobility had taken a backseat to military prowess and organizational skills in the selection of retinue commanders. In the background, freemen and their families also labored at husbandry and agriculture. In Caesar’s time the Germans largely were pastoralists but Tacitus describes peoples more focused on sedentary agriculture. Freemen also supplied labor for military action and raiding.

The Gauls to the west of the Rhine generally were wealthier than their German neighbors; raiding opportunities thus were tempting to begin with. Those opportunities expanded greatly after Caesar’s conquest. Maintaining Rome’s frontier forces became big business. Along the frontier, food, supplies, and money became available for the taking in hitherto unknown quantities. Tacitus describes a Germania wherein the returns to raids had risen substantially. Consistent with this conclusion, Heather (2010, pp. 139–141) relates the illustrative example of Vannius, a king of the Germanic Marcomanni confederation in the first century AD. Vannius became a frontier client of the Roman Empire, recognizing “the wealth-generating potential of making Germanic traders bring their goods to Roman merchants on his soil, so that he could charge tolls”. However, in 50 AD his own wealth was expropriated by another group of Germans from beyond the frontier zone.

Rome was encroaching upon areas that had also served as pressure valves for the growing populations of the relatively (to Gaul and Rome) pastoral Germania. Much of Germania was covered by dense forest that was costly to clear. Facing scarcity of land suitable for husbandry and pastoral agriculture, Germanic peoples often chose to migrate westward. The establishment of the Roman frontier raised the costs of westward migration. Rome sought to regulate the settlement of barbarians within its provinces and it stood ready to protect existing settlements within its borders. As such, the relative returns to sedentary agriculture increased. Moving to sedentary agriculture provided relief from the demand for new lands. “This transformation, then, from a preeminently pastoral state to an agricultural state came about during the century after [the Germans’] expansion was arrested by the power of Rome” (Bury 1967, p. 7).

Expansion to the east was, of course, an alternative to butting heads with Rome. Germania stretched eastward about 500 miles from the Rhine through what is modern-day Poland. Not coincidentally, the eastern limit of Germania corresponded almost precisely to that of the North European Plain, a vast expanse of fertile lowlands. Eastern expansion would have brought Germans into lands less-than-ideal for sedentary agriculture. Furthermore, Germans expanding to the east would encounter non-Germanic tribes: the Venedi to the north and Dacians to the south, as well as Iranian-speaking nomads. In particular, the periodic migration of new nomadic groups from the Asiatic steppes over the Black Sea was characteristic of the region, causing trouble both for the Romans along the Danubian frontier and the eastern Germanic peoples. These migrations included the Alans in the first century BC and then notoriously the Huns in the fourth century (Todd 1987, pp. 37–38). Alternatively, expansion to the west involved dealing with settled and relatively stable communities.

Since I argue that the encroachment of Rome strengthened Germanic incentives towards sedentary agriculture, it is appropriate to here address two occasions when Tacitus alludes to Germanic laziness, in part with respect to agriculture. Given my reliance on Tacitus’ observations of governance institutions, I need to make clear why these allusions to laziness should be discounted. First: the German “thinks it tame and spiritless to accumulate slowly by the sweat of his brow what can be got quickly by the loss of a little blood (Tacitus Chap. 14, pp. 113–114). Second: “The fact is that although [the Germans’] land is fertile and extensive, they fail to take full advantage of it because they do not work sufficiently hard” (Tacitus Chap. 26, p. 122). While one must be wary of picking and choosing which of Tacitus’ statements to take seriously and which to dismiss, it is accepted widely that Roman ethnographers “employed a dichotomy between civilized and uncivilized, urban civilization and barbarians, as a basic tool in their analyses” (Burns 2003, p. 3). For Tacitus, this dichotomy served as a mirror by which Romans could examine their own morality—he low points and the high:

The Germans are free of the vices or sophistication and cities. They are nonetheless barbarians, indolent, slothful, excessive drinkers, in the main wanting in direct intelligence. Like all barbarians, they cannot withstand unexpected trials and are fit only for sudden exertions, not for sustained labor. (Fitzsimons 1976, p. 478)

Through his portrayal, Tacitus hoped that the Germans’ “customs may inspire Romans to return to their own earlier austerity and rigor […]” (Fitzsimons 1976, p. 478). Caesar undoubtedly employed a more or less similar dichotomy. The distinctions between Tacitus’ and Caesar’s portrayals of Germanic governance institutions are likely to be more salient than the similarities.

Furthermore, when Tacitus notes that the German’s “land is fertile and extensive” but that they do not “take full advantage of it” owing to a lack of effort, his comments may refer to farming itself or to the labor necessary to clear arable land in the first place. The latter would be consistent with archaeological evidence pointing to the growing importance of agriculture at the turn of the millennium. Todd (1987, p. 84) surveys evidence from an excavated settlement located in coastal marshes between the Weser and Elbe rivers:

In the late first century BC, the marshland […] first became amenable to settlement. Arable farming is revealed in rough plough marks in the natural soil at this time […]. No traces of buildings have been recovered from the beginnings of this phase. Near its end however, two substantial houses and a granary were constructed, and by the first century AD, seven or eight dwellings […] and their attendant granaries, had appeared.

Todd (1987, pp. 87–88) states further that

No doubt before the marshes were actually settled they had been used for fishing, the catching of game, and perhaps as pasture grounds for cattle, but the determined onslaught upon them in the early Roman period was clearly made by settlers seeking new land […].

In Todd’s (1987, p. 88) opinion: “The colonization of the marshland may [have been] an early indication that the land suitable for agriculture in these parts of Germania would one day prove inadequate for the needs of its population.” This archaeological evidence, of course, is not amenable to precise dating; the evidence also is limited to excavated sites. Again, however, Todd (1987, Chap. 3) discusses a number of examples, including the Weser-Elbe marshlands settlement, that are generally consistent with the expansion of sedentary agriculture in the face of population pressures exacerbated by Rome’s encroachment.

From Caesar to Tacitus, then, Germanic peoples faced stronger incentives to reallocate their resources and efforts away from pastoralism and towards both sedentary agriculture and raiding across the frontier into the Roman provinces. Caesar (book VI, p. 347) tells us that the leading men assigned land to tribes and clans (gentibus congnationibusque hominum). A system of annual land reallocation to tribes and clans, rather than individuals, may have worked well for husbandry. However, it may also have been an inefficient arrangement for sedentary agriculture. The internalization by individual farmers of the benefits over time to fertilization and crop rotation is relatively important. Likewise, to the extent that raiding became more important, being able to choose commanders based on organizational skill and military prowess rather than their noble birth would be important.

Notably, “Tacitus makes it clear that only military leadership had survived among those closest to Rome and that only those farthest away were still ruled purely by sacral kings” (Burns 2003, p. 180). This statement indicates how Rome’s encroachment was the source of larger returns to raiding activities. The importance of those raiding activities increased especially for barbarian groups who were closer to the Empire’s frontiers. Furthermore, movements towards sedentary agriculture and raiding were complementary to the extent that “the massive increase in food production that this revolution in agricultural production must have generated goes a long way towards explaining how the new military kings could support their retinues” (Heather 2010, p. 51).

The changes described above would work more to the benefit of the freemen than the nobility. The freemen provided the labor for the production of both food and violence. The nobility, alternatively, stood to lose its exclusive role in military leadership and its control over the allocation of land. The nobility were not the primary source of labor and were outnumbered vastly by the freemen. Relinquishing military and land-allocating roles would leave the Council of Leading Men without a well-defined part to play in Germanic society, and the (no longer exclusively noble) duces might render the reges irrelevant.²⁴ The expanded range of freemen decision-making would impose external costs on the nobility. In particular, nobles would have no say in either the decision to go to war or its execution once begun.

While the losses suffered by the nobility following these constitutional changes were significant, the gains to the freemen were large enough to offset them. Congleton (2007, p. 270) notes that, within the King and Council template, “reform is possible whenever circumstances change and the [council] or the king can fully compensate the other for anticipated losses from new procedures and constraints”. In this context, the reforms to land allocation procedures and the constraints imposed on military election were made possible by rewards offered to the Council of Leading Men by the Assembly of Warriors. This compensation took the form of agenda control in a regularly convened Assembly.

Unfortunately, Tacitus does not describe the “major affairs” of the assembly in detail. Aside from law and order, the only public goods explicitly referred to are horses “kept at public expense” for the purpose of announcing omens and prophesies to the community (Tacitus, Chap. 10, p. 110). However, in the provision of law and order Tacitus notes: “The man who is found guilty has to pay a fine of so many horses or cattle, part of which goes to the king [regi] or the state, part to the victim of the wrongful act or to his relatives” (Tacitus Chap. 12, p. 111; emphasis added). This passage suggests that fines served in lieu of taxes. Furthermore, the Assembly chose “magistrates who administered justice in the districts and villages” (Tacitus Chap. 12, p. 111). The magistrates presumably had significant influence in determining the amount of fines that were collected, and perhaps what share of them went to “the king or the state”.²⁵ The following statement by Tacitus (Chap. 15, p. 114) is noteworthy: “It is a national custom for gifts of cattle or agricultural produce to be made to the chiefs [principibus], individual citizens making voluntary contributions for this purpose”. If “voluntary” here refers to the unanimous decision by the freemen of the Assembly, then this again suggests that “voluntary contributions” substituted for taxes and were sources of income for the nobility.

Rome’s encroachment upon Germania effectively enlarged the potential sources of wealth available to barbarians. The constitutional reforms that occurred provided them with greater access to those potential sources. On the one hand, the election of retinue commanders based on organizational skills and military prowess allowed freemen to more effectively pursue opportunities for wealth extraction, in particular those opportunities associated with raiding across the frontier. On the other hand, by administering fines and the collection of voluntary gifts from the community the nobility stood to benefit. With an expanding wealth base, even holding the line on fines and gifts meant that the principes were better off. Moreover, fines were associated with the adjudication of torts and crimes. The constitutional exchanges provided for more regular and frequent opportunities for the nobility to hear and adjudicate charges.

In addition to the Assembly of Warriors meeting monthly, the Council of Leading Men set its agendas. As Congleton (2007, 2011a) demonstrates, agenda control is particularly valuable in uncertain environments that are subject to changes in policy preferences. Rome’s encroachment created just such an environment for the Germans and their nobility in particular. Agenda control allowed the nobility to preserve the status quo (by not allowing alternative policies to be considered) or to restrict the alternatives under consideration. Those alternatives would be restricted to those leaving the nobility better off or at least no worse off.

In achieving the constitutional exchanges with the nobility, the freemen may have faced commitment problems (Acemoglu 2003). For example, the Council of Leading Men relinquished control of land allocations. Individual freemen were left in a better position to exercise “exit options” and avoid the actual collection of fines. How these commitment problems were overcome is an interesting question. In the case of exit options, the encroachment of Rome and population pressures likely helped. Individual freemen faced tighter constraints on their ability to “vote with their feet”. This would have made the acquisition of agenda control in the Assembly of Warriors more meaningful to the nobility. Magistrates could have more easily administered justice and collected fines, securing the revenues due to the nobility.

6 Concluding discussion

In this paper I have provided an economic analysis of Germanic barbarian institutions of self-governance during and their evolution from the first century BC to the first century AD. The analysis makes the observed changes intelligible in terms of constitutional exchanges between the Germanic elite and freemen. These exchanges were motivated by new sources of wealth and greater scarcity of land, both linked to Caesar’s conquest of Gaul.

This paper contributes to the literature on self-governance. That literature has been in large part limited to studying pockets of anarchy embedded within or adjacent to actual governments. Archaeological evidence suggests that “the lands east of the middle and lower Rhine had been occupied by peoples in small communities, and there is no evidence of interregional political organization” (Wells 1999, p. 233; also see Chap. 3). Tacitus also names about 40 different Germanic tribes inhabiting the vast geographic area of Germania.²⁶ The institutions of self-governance described above appear to have generally characterized these barbarian groups.²⁷ Those institutions were not embedded within a larger governance framework.

First century Germanic self-governance may speak to the question of whether or not self-governance can be effective across large geographic areas in the absence of government as conceived of in modern times. This question is relevant to studies of the anarchic environments associated with failed states (e.g., modern-day Somalia: Leeson 2007c; Powell et al. 2008). In such environments, individuals may face governance options that do not include a non-predatory state (Leeson and Williamson 2009). This may lead them rationally to choose self-governance over government on efficiency grounds (Leeson 2007a). What were the alternative governance options available to the Germans? How effective was self-governance relative to those alternatives? These are difficult questions. Regarding the former, the Roman Empire was unlikely to absorb Germania (Sect. 1 above). Germanic tribes could migrate across the frontiers but doing so was costly. Whether or not the Romans would permit settlement and under what conditions also were very uncertain matters. We also have to ask whether one would classify the Roman Empire as a predatory state.

Regarding the effectiveness of the barbarians’ self-governance, we know that their constitutions effectively adapted to the changes in economic conditions that followed Caesar’s conquest of Gaul. We would like to know more. How were intertribal relations managed and disputes resolved? Was intertribal violence frequent or was peaceful coexistence the norm? Relative to what benchmark? These are interesting questions for future research. However, answering them will require the harnessing of not only the (very scant) evidence of first century barbarians, but also the (not abundant) evidence of barbarians from the later Roman period. Harnessing that evidence is beyond the scope of this paper.

First century Germanic barbarians also provide an example of self-governance across different social groups (Leeson 2009b). In the century following Caesar’s conquest of Gaul, mutually beneficial constitutional exchanges occurred between the freemen and elite. As part of these exchanges, the elite relinquished exclusive claim to military leadership positions in favor of a meritocracy of professional commanders. This occurred through exchange rather than violent conflict between the two social groups.

Furthermore, different roles for the adjudication of disputes were established between the Assembly of Warriors (all freemen) and the Council of Leading Men (the elite). This Germanic system of adjudication presumably provided for order across not only freemen and elites, but also across the many competing and potential hostile armed retinues. This is remarkable not only because the Assembly and Council lacked a monopoly of force. Germanic self-governance was constituted such that the weaker organizations were able to provide order across stronger organizations specializing in the provision of violent services (Leeson 2007d).

Among others, Friedman (1979), Anderson and Hill (2004), and Leeson (2007b) have documented the emergence of decentralized legal systems within social groups. However, along with Leeson’s (2009c) study of the Anglo-Scottish borderlands, the Germanic barbarians represent a rare example of legal self-governance operating across different social groups. Many scholars have expressed the view that self-governance is ineffective across social groups (e.g., Landa 1994; Zerbe and Anderson 2001; Grief 2002; Dixit 2004). The experience of the Germanic barbarians also calls this prior into question.