Download file in PDF format: TRAC 2007: Developing the ‘Germani’ in Roman Studies (pp. 131–150)
Developing the ‘Germani’ in Roman Studies
Cheryl Louise Clay
Introduction This paper explores the significance of the term ‘Germanus’ (German) in reference to Germanic tribes from the Rhineland, in particular those from the provinces of Germania Inferior and Superior. In contemporary scholarship the term is treated as a macro-label forced upon tribes by outsiders, which had no emic (self-ascriptive) significance to the tribes themselves since it is maintained that they perceived no pan-tribal sentiments. An under-evaluated source of evidence, however, is the epigraphic record from Rome and Britain, where the term is used as an emic definition. This paper explores these inscriptions and discusses what contribution they can make towards re-evaluating the identity of the Germani in Roman studies.
Outlining the Problem The term ‘Germanus’ receives cursory attention in most academic publications. The reason is that it is perceived as a Roman construct whose closer inspection yields little further assistance in understanding the identities of former Germanic groups. The term is considered to have held no currency amongst the tribes themselves since they entertained no such pan-tribal, affiliations. The following lists some of the most definitive statements made in leading publications. ‘The Germani….had no collective consciousness of themselves as a separate people, nation or group of tribes. There is no evidence that they called themselves ‘Germani’ or their land Germania.’
‘All that is reasonably certain is that a member of a German tribe, when asked about his or her affiliations, would have answered Langobard, Vandal, Frisian or Goth, not Germanus.’ Todd (2004: 8 and 9) in The Early Germans.
‘The peoples surveyed by Tacitus or those of the Migration Age were fragmented; they did not call themselves Germans but bore particular names.’ Goffart (2006: 5) in Barbarians Tides.
‘Attaching archaeological substance to historically transmitted ethnic macro-concepts like ‘Gauls’ and ‘Germans’ is also fraught with risk, given that they were largely Roman creations that had little value as self-ascriptive, emic concepts for individuals or groups.’ Roymans (2004: 29) in Ethnic Identity and Imperial Power: The Batavians in the Early Roman Empire.
The word Germanus is also generally understood not to have been employed in the epigraphic record by the persons themselves and Maureen Carroll (2001: 113) in Romans, Celts and Germans. The German provinces of Rome, for instance, states:
‘inscriptions recording the origin of Germanic groups throughout the Empire consistently cite the tribal affiliations of the persons in question.’
This type of opinion towards the Germani has been little affected by either processual or postprocessual approaches and it has stood resolute for more than the past thirty-five years:
‘The peoples whom the Romans called Germani did not apply this name to themselves: indeed they had no single term comprehending all Germani. They had no consciousness of any ‘Germanic’ ethnic community, recognising only a variety of separate tribes, large or small; and in referring to themselves they used the tribal name’. Hachmann (1971: 16) in The Germanic People.
The problem with this type of approach is that on the inscriptions from Rome and Britain the term ‘Germanus’ is used as an emic definition by persons originating from the Rhineland. Whilst some scholars are familiar with the British examples, they are so rarely discussed in academic literature that archaeologists dealing with Germanic tribes on the Continent barely appreciate that these inscriptions exist. This corpus of evidence has, therefore, been unavailable for theoretical discussion or for comparative analysis alongside the material from Rome. The intention of this paper is to present this evidence for consultation in order to stimulate a wider debate. The Lower Rhine exhibits one of the poorest densities of stone inscriptions from the entirety of the Roman world (Woolf 1998, 83), and only a few of these were produced at the behest of ‘natives’. We are more fortunate in the fact that the natives from these communities were more inclined to make written reports of their own ethnonyms when they were operating outside this domain in more epigraphically dense polities and provinces (Carroll 2001: 128; Roymans 2004: 232; Derks 2004). The provinces peripheral to the German provinces thus provide important resources for quarrying epigraphic information written by the groups themselves. Whilst the perceptions of migrants may differ from those still residing in their own territories, these texts provide some of the finest exploratory interfaces for studying the transmission of native identities within a Romanised milieu. Inscriptions are perceived as attempts ‘to assert identities’ (Woolf 1998: 78), and are ‘linked intimately to the expression of identity’ (Carroll 2007: 28), and can be compared with the etic (external) uses given in the Classical sources.
Literary Background to the Germani and the Two Germanies The premise that the term ‘Germanus’ held no significance to the tribes themselves has never been fully substantiated by the documentary sources; indeed they have always alluded to the contrary. Tacitus claims that the Nervii were proud of their Germanic heritage despite the Romans’ reservations about their origins (Germania 28), whilst the Ubii were supposedly chastised by neighbouring communities for increasingly distancing themselves from their German status (Tacitus Histories 4.28). This suggests, that whatever the etic perceptions of Roman writers, the communities in question retained their own opinions about the appropriate application of the term Germanus, which occasionally contradicted the Roman view.
The earliest surviving passages, which provide a geographical description of the Germanic tribes, are found in Caesar’s Bellum Gallicum dating to the mid-first century B.C.. Here the Germani are identified as residing predominantly north of the Rhine, but some of the groups had also migrated onto its lower, right bank and were referred to as the Germani cisrhenani – ‘the Germani on this side of the Rhine’ (Bellum Gallicum 2.3.4; 2.4.10; 6.2.3; 6.32.1; Figure 1). Caesar goes into further detail about who these tribes were designating the Eburones, Segni, Condrusi, Paemani, Caerosi, Cimbri and Teutons as part of this group (Bellum Gallicum 2.4.10; 2.29.4; 6.32.1).
The etymology of the term ‘Germanus’ remains somewhat obscure (Todd 2004: 9), and Strabo (Geography 7.2) assumed that it was a Roman appellation derived from the Latin ‘Germanus’ meaning genuine. That a Germanic tribe may have been originally responsible for coining the word, however, is supported by the stem ‘-manus’ utilised in other tribal names of suspected indigenous origin such as the ‘Marcomanni’, ‘Paemani’ and ‘Alammani’. Tacitus (Germania 2) also confirms that Germanus was a native coinage, which received greater exposure during the time that certain east-bank tribes were migrating into Gallic territory: ‘The first tribes in fact to cross the Rhine and expel the Gauls, though now called Tungri, then bore the name Germans. So little by little the name…prevailed until the whole people were called by the artificial name of Germans’.
Figure 1. The distribution of the Germani cisrhenani in the first century B.C.. Roymans, N. 1990. Tribal Societies in Northern Gaul: figure. 2.1 page 13.
That the notoriety of one group could extend to a wider movement of people is a common ethnographic phenomenon, which seems to have occurred in the case of the Alemanns and the Saxons as well as the Germani (Wenskus 1961: 99). Recent approaches towards ethnicity also propose that ethnic groups are most likely to be forged in the context of opposition to other societies (Jenkins 1997: 13; Jones 1997: 45), which is, naturally, compatible with Tacitus’ passage. ‘Germanus’ may have been initially adopted by a number of tribes partly through contact with, and opposition to, indigenous ‘Gallic’ groups.
From the mid first-century B.C. onwards many of the afore-mentioned Germanic tribes were supplemented and superseded by other transrhenine groups penetrating the Rhine frontier (Fig. 2). In many cases Rome facilitated their entry in order to stabilise this frontier against more hostile incursions from across the Rhine. The Batavians, Ubii, Cugerni and Vangiones were all originally allowed admittance specifically in order to participate in military service (Carroll 2001: 29–31; Roymans 2004: 24, 58). Recruitment from amongst many of the Rhineland’s tribes came to be socially significant, precisely because its frontier always demanded a great many troops (Schönberger 1969; Mertens 1983). Some of the earliest armies raised from these tribes were simply referred to as the ‘Germani’ (Roymans 2004: 56), whilst estimates predict that at least one male from every family on the Lower Rhine would have been mobilised into military service (Derks and Roymans 2002), and recruitment from the Upper Rhine was also significant. In A.D. 83–90 the provinces of Germania Inferior and Superior were created out of these militarised frontiers, with the implication that the geographical appellation of the Germani was extended to encompass the entire vicinity of the Rhineland (Carroll 2001: 40; Roymans 2004: 196).
Figure 2. Outline of the provinces of Germania Inferior and Germania Superior with circles around the principal tribes mentioned. Arrows indicate tribes whom were historically recorded to have migrated from the east-bank of the Rhine. Adapted from Carroll, M. 2001. Romans, Celts and Germans. The German provinces of Rome: figure 4 page 30.
Initial Languages spoken by Germanic migrants The linguistic situation regarding Germanic migrants from across the Rhine has always been controversial but studies suggest that some might have spoken a language that is ancestral to the Germanic languages spoken today. In a similar fashion to the Romance languages of Italian, French, Romanian and Spanish, all of which are the descendants of Latin (see Bonfante 1998: 3), the current Germanic as well as Celtic languages are the offspring of languages that were once spoken within and beyond the Roman empire. It is largely beyond dispute that in the early Roman period Celtic languages were still spoken throughout most parts of Gaul, Britain and Spain (Whatmough 1970: 37, 665; Gregor 1980: 6; Rankin 1997: 8–26; Collis 2003: 128–132; Raybould and Sims-Williams 2007). By contrast Germanic had originated in northern Germany but it had spread to many regions east of the Rhine (Polomé 1983, 527–531; Nielson 1998, 40). Archaeological evidence supports the immigration of groups onto the right bank of the Rhine documented in the historical sources (Gechter 1990; Wells 1999, 82–3; Carroll 2001, 126; Roymans 2004: 27, 118–130, 205). It seems that a Germanic language was imported at the same time. Caesar (Bellum Gallicium 1. 47), Tacitus (Germania 43) and Suetonius (4. 47) all refer to Germanic tribes speaking native languages which differed from those of the Gauls, and the philological evidence for the Germanic-ness of this language is provided by the names of persons, deities and places recorded on the Rhineland frontier (Weisgerber 1968: 143, 150; Schmidt 1980; Neuman 1987). The Ubii, Batavi and Cannanefates are argued to have been native Germanic speakers (Weisgerber 1968, 118, 131, 143; Toorians forthcoming). Place-names, such as the ‘insula Batavorum’ (Bellum Gallicum 4. 10) dating to as early as Caesar’s ‘Germani cisrhenani’, as well as the later ‘vada’ (Tacitus Histories 5. 21), contain the Germanic etymologies of ‘good island’ and ‘wade’ respectively (Toorians forthcoming). Thus, it seems plausible that whilst Celtic was the dominant linguistic amongst much of Roman Europe during the initial phases of its conquest, some of the earliest Germanic migrants spoke dialects that were related to that linguistic group which now takes their name. It is also believed that during the course of the Roman period, a certain number of Germanic words (such as ‘burg’ fortress and ‘wardon’ to ward) were adopted into Vulgar Latin through the context of German-speaking recruits within the Roman army (Elcock 1960: 214–24; Bonfante 1998: 79; Adams 2003: 274–279). This is more evidence to suggest that Germanic dialects were introduced to the Rhineland, and it shows that some of them could receive a wide exposure under a Roman milieu.
The Batavian Revolt and the Germani in the Rhineland and Rome The so-called Batavian Revolt is infamous as an indicator of pan-German sentiments amongst the tribes situated in the Rhineland. This Revolt was first roused in opposition to the government’s overly zealous demands for native troops to be incorporated within the Roman army (Tacitus Histories 4.13–14). The peoples of the Batavi, Tungri, Cananefates, Frisians, Vangiones, Chatti, Chauci, Mattiaci, Bructeri, Tencteri, and the Cugerni were all involved in this event (Tacitus Histories 4.15–17, 27, 32, 37; 5.16, 19), and one of the most unequivocal expressions of their pan-tribal, ‘German’ sentiments is put into the mouths of the Tencteri when they attended a native assembly held by the Ubii in Cologne. As previously mentioned, the Ubii had been once chastised for distancing themselves from their ‘German’ status, but through the convention of such an assembly the Tencteri declared them to ‘have been returned to the body of the German people and to the German name’ (Tacitus Histories 4.63).
As early as the reign of Augustus, some of the members of these Rhineland groups had been enlisted to serve exclusively in the emperor’s bodyguard stationed in Rome (Suetonius 2.49; 4.43.). This practice continued until the reign of Nero (A.D. 54–68), but the bodyguard was disbanded at the time of the Batavian Revolt due to the state’s concerns over these soldiers’ loyalty (Tacitus Histories 2.5; Suetonius 4.12). Memorials set up by the members of this guard date from the time of Claudius (A.D. 41–54) to its disbandment, which naturally provide important records about the self-perceptions of soldiers recruited from these native tribes up until this critical point in the Rhineland’s history. Fifteen inscriptions give the natio (nation, tribe) of the deceased. Ten of these use the term ‘Batavus’, three belong to the ‘Ubii’ (from the region of Cologne), two use ‘Suebi’ and one ‘Baetasi’ (Bellen 1981: 36). Modern studies often use these monuments to highlight the importance of tribal definitions in the construction of the soldiers’ native personas. Roymans (2004: 28), for instance, states ‘in their grave inscriptions….. they emphasised only their tribal identity’. It should not escape our attention, however, that the guild to which these soldiers belonged was entitled the collegium Germanorum – the society of the Germani (Fig. 3). This is in reference to their own burial club which soldiers themselves were generally responsible for both founding and naming (see Goldsworthy 2003: 50). Such societies did not exist simply in order to bury the dead but they also played an active role in the lives of the living who would use such societies to hold regular meetings and to partake in collective worship (Bellen 1981: 60). The pan-tribal nature of this German collegium is clearly identified on the basis of the tribal signifiers of ‘Batavus’, ‘Suebus’ and ‘Ubius’ belonging to this society. Twenty other inscriptions recovered from Rome also cite ‘Germanus’ as an emic definition (Noy 2000: 216; Fig. 4), which in any case undermines the premise that only tribal ethnonyms were employed by these soldiers. The groups recognition of the name of the Germani is even more edifying when one remembers that these records date to before the creation of the provinces of Germania Inferior and Superior. During this period the Rhineland still loosely fell within the jurisdiction of Gaul (Pliny Natural History 4.17.106). It is often assumed that any native application of the term ‘Germanus’ must be ‘imitating Roman usage’ (see Wolfram 1997: 4), but when the inscriptions broadcasting the German-ness of these tribes precede the literary sources, one is tempted to wonder whether the evidence has been interpreted in the right order? For instance, Strabo (64 B.C.–24 A.D.) in a well-known passage of his Geography (4.4.2) refers to the entire communities of the lower Rhine including the ‘soldiers’ (belonging to what would have then been the Batavian polity) as ‘now called Gauls’. One suspects that the Batavian Revolt played a fundamental role in re-negotiating the Romans’ understanding of tribes situated on the Rhineland frontier. Surely, the inscriptions set up in the Imperial capital must also have contributed to the manipulation of such external perceptions. Memorials were designed with the objective of shaping the mentalities of their intended audience (Carroll 2007: 53–55). Consequently, the inhabitants of Rome would have needed to look no further than their own roadsides in order to gauge an impression of the soldiers’ relationship with the Germani. This is not to suggest that groups considered themselves as Germans in a nationalistic sense, however, and the reasons why some peoples may have considered themselves ‘German’, as well as who these people were, will at least be partially explained in the context of the inscriptions that have for long been uncovered in Britain.
Figure 3. Tombstones from Rome dedicated to the memory of soldiers attached to the collegium Germanorum. Each soldier belonged to a different tribe. From left to right: ‘natio Batavus’, ‘natio Baetesius’ and ‘natio Ubius’. Deutsches Archäologisches Institut Rom, Neg. Nr. –78. 465, Nr. –78. 470 and (far right) Heinz Bellen.
Figure 4. One of twenty tombstones from Rome whereby Germanus is cited as an emic definition. Deutsches Archäologisches Institut Rom, Neg. –Nr 77. 1601.
The Germani on the Hadrianic Frontier The epigraphic evidence for the Germani in the vicinity of Hadrian’s Wall dates to the second and third centuries and is a by-product of the considerable number of auxiliary and irregular units who were stationed at many of its installations (Clay 2007). The term ‘Germanus’ is found in twelve very different inscriptions, which do not conform to a particular society or tight-knit epoch so that each text contributes to an incremental development of the image of the Germani. Most of these emic references occur on altars rather than memorials, which contrasts from the case at Rome, but must be seen in the context of the relatively few tombstones that have been uncovered in general from the northern frontiers of Britain (Fig. 5).
Figure 5. Distribution of inscriptions from Hadrian’s Wall where the term Germanus is cited as an emic definition. Author’s own.
The Germani at Carrawburgh The first site to be discussed is the water sanctuary, which in the early third century had been built by the First Cohort of the Batavians to the local goddess Coventina (see Allason-Jones and Mckay 1985: 11). A conspicuous feature of the temple is that in addition to the Batavian unit, the other altars had either been set up by members of different Germanic auxilia or by persons with non-local, ‘Germanic’ personal names (Werle 1910: 33, 45), more likely originating from the Rhineland than Britain. Two monuments were erected by tribunes of the First Cohort of the Batavians (RIB 1534, 1535), but another altar was raised by a commanding officer of the Cuberni (RIB 1524), and a fourth inscription by the First Cohort of the Frisiavones (RIB 1523). On two of the altars dedicated by persons with suspected Germanic personal names, the dedicators ‘Crotus’ and ‘Maduhus’ profess themselves as ‘Germanus’ (RIB 1526, 1528). In addition to not providing tribal signifiers, as was supposedly the protocol, some of the members of this religious community appear to have observed a communal consciousness of themselves as ‘Germans’. This parallels the types of relationships already noted in Rome (Fig. 6).
On the basis of the negative terminology about the Germani found in Roman literary sources, it has been assumed that associations with persons classed as Germans could not have been favourable (Roymans 2004: 28). Whilst this seems acceptable, the prejudices of the patrician and equestrian orders may not have been representative of the opinions maintained by the public in general. The well dedicated to Coventina is unusually wealthy, yielding more than 160, 000 coins, which represents the largest cache of its kind uncovered from Northern Britain (Allason-Jones and McKay 1985: 50). Here, Germanic associations were clearly not an impediment in either the economic or religious sphere. Presumably, there might have been a plurality of perceptions towards the Germani, but not all would have been either negatively imposed or received.
COMMUNAL WORSHIP AT CARRAWBURGH Crotus Germanicus First Cohort of the Frixiavones First Cohort of the Cubernians First Cohort of the Batavians Maduhus Germanicus Figure 6. At Coventina’s Well, Carrawburgh members from three different Germanic auxilia are attested by the epigraphic record as well as two persons who profess themselves to be Germanus. Author’s own.
The Germani at Chesters, Old Penrith, Ebchester and Brampton In the second century A.D., the fort at Chesters was occupied by the First Cohort of the Vangiones (Holder 1982: 123). From this fort we find one memorial and one altar with the term ‘Germanus’. A certain ‘Venenus Germanus’ had installed the altar within the fort’s bathhouse (RIB 1449). The positioning of the monument in this prominent location, employed for social gatherings, might betoken the same enthusiasm, or even pride, attached to the term, which has already been seen at Coventina’s sanctuary.
The memorial found on the outskirts of this fort was erected by a certain ‘Lurio Germanus’ to his wife Julia, sister Ursa and son Canio (RIB 1483; Fig. 7). The inscription allows a more biographical insight into persons being referred to as Germans, since neither Lurio’s sister nor his son are defined with the term. This may suggest that the status was restricted by both gender and lineage, which is corroborated by the other examples. A memorial at Old Penrith, for example, was set up by Vindicianus to his brother ‘Crotilo Germanus’, as well as his sister ‘Greca’. This shares the same pattern in that the term is not used for female relations. However, here, even the brother Vindicianus denies himself the label (RIB 934). It is probable that the two brothers would have been born in the same area, which suggests that a German status was not simply defined by geography.
On two altars from Ebchester (RIB 1102) and Maryport (RIB 841), ‘Virilis’ and ‘Labareus’ each refer to themselves as Germanus whilst paying homage to their respective deities of ‘Vernostonus Cocidius’ and ‘Setlocenia’. Virilis is the only German recorded on the frontier to possess a Roman cognomen (meaning ‘manly’ in Latin). The altar set up by the four Germani ‘Durio, Ramio, Trupo and Lurio’ at Brampton (RIB 2063; Fig. 8) dates to the second century and similarly lists persons without Roman citizenship and names not belonging to Latin nomenclature. These Germani can perhaps also be identified with persons recorded on other inscriptions. One wonders, for instance, whether the ‘Lurio’ recorded alongside his colleagues at Brampton is the same ‘Lurio Germanus’ who set up the memorial to his family at Chesters? Another ‘Thrupo’ with a variant spelling from the Brampton ‘Trupo’ is also identified as a centurion at the fort of Carrawburgh (RIB 1556). If these two latter inscriptions actually represent the same person they provide an insight into the military background behind somebody referring to himself as Germanus, where all other inscriptions have remained elusive in this respect. The Carrawburgh ‘Thrupo’ was not an ordinary soldier but a commander of other men. In any case, the fact that the four men at Brampton embraced the term Germanus shows that they were emphasising similarity between each other rather than highlighting tribal difference.
Figure 7. Memorial from Chesters by Lurio Germanus to his wife Julia, sister Ursa and son Canio. Collingwood, R. G. and Wright, R. P. 1965. Roman Inscriptions of Britain I: catalogue number 1483 page 477.
Figure 8. Altar from Brampton dedicated by four Germans Durio, Ramio, Trupo and Lurio. Reproduced by permission from Corpus Signorum Imperii Romani. Corpus of Sculpture of the Roman World. Great Britain, vol. 1, fascicule 6: number 158. The British Academy.
Figure 9. First-century tombstone from Cirencester dedicated to the memory of a soldier from Frisia (civis Frisiaus). Collingwood, R. G. and Wright, R. P. 1965. Roman Inscriptions of Britain I: number 109 plate 4.
‘Germanus’ is also utilised in the epithets of two deities worshipped on Hadrian’s Wall. This betokens a cosmological significance to the label and in one instance derives from another Germanic unit, who was in some way connected with the term. The ‘Matronae Germani’ are invoked on a rather damaged inscription (RIB 2064), but the goddess ‘Garmangabis’ (RIB 1074) is celebrated by a vexillatio of the Suebi. It is possible that ‘Garman’ reflects a variant orthography for ‘German’, which is found in other records, albeit dating much later (see Bede Ecclesiastical History 5. 9). Unless this Suebian unit was mobilised from beyond the empire, it must have been recruited from the ‘civitas Sueborum Nicrensium’ situated in the Agri Decumates of Upper Germany (see Drinkwater 1983, 61; Wigg 1999; Sommer 1999).
The Germani at Housesteads, and their Implications Finally we turn to the Germani at Housesteads. Two altars pertaining to the Germani have been recovered. By the early third century this fort was occupied by the ‘cuneus Frisiorum’ – a Frisian fighting unit (RIB 1574), and the ceramic evidence recovered from the site also points to migrants from the Texel region of the Netherlands dating to the same period (van Driel-Murray forthcoming). The Frisians were situated north of the Rhine during the Roman period, but they were another group routinely raised into the Roman army from an early period, and a first-century memorial from Britain records such a soldier from Frisia (ciuis Frisiaus) (RIB 109; Fig. 9).
A small, portable altar recovered from within the fortress at Housesteads had been dedicated by a certain ‘Calve Germanus’ (RIB 1597). Although somewhat weathered and poorly carved, the deity recorded was probably M(ars), whose popularity amongst the Germans at Housesteads is further attested by a temple on the fortress’ outskirts, which had been dedicated to ‘Mars Thingsus’ by the ‘cives Tuihanti Germanorum’ (RIB 1593). The temple dates to the early third century, and another altar recovered from this same site reveals that the ‘cives Tuihanti’ had been allocated to this ‘cuneus Frisiorum’ stationed at Housesteads (RIB 1594; Fig. 10). Based upon place-name evidence the ‘cives Tuihanti’ have been traced to the region of Twente in Overrijsel, which was recorded as ‘pagus Tuianti’ in A.D. 797 (Künzel 1988: 262). As with the Frisians, then, this was another tribe recruited from beyond the Rhine frontier into the Roman army. Despite the generic associations of the term ‘cives’ (meaning ‘people’) it is likely that the inscription referred only to the cives’ men, since it is difficult to envisage women participating in a Frisian fighting unit and, as already noted with other inscriptions, the term ‘Germanus’ has been denied to women and offspring.
Figure 10. From the temple dedicated to Mars Thingsus, Housesteads one of the altars recorded that the ‘cives Tuihanti’ had been allocated to the ‘cuneus Frisiorum’, which was stationed at the nearby fort at Housesteads. Collingwood, R. G. and Wright, R. P. 1965. Roman Inscriptions of Britain number 1594 page 508.
Excavations revealed that the temple consisted of a circular arena four metres in diameter, from which three altars were recovered (Birley 1962: 121; Figure 11). The altar set up by the ‘cives Tuihanti Germanorum’ served as one of two lintels forming an entranceway to this sanctuary supporting an arch (Figure 12). The other altar/lintel has not been discovered, but the arch depicts Mars Thingsus armed with a shield and spear. The dramatic appearance of this entrance-way to its members can be well imagined and the usage of the term ‘Germanus’ would have been an integral component of its visual and written iconography. The three altars recovered from the site also imply the participation of three distinct Germanic groups: the ‘cives Tuihanti’ (RIB 1593; 1594), ‘cuneus Frisiorum’ (RIB 1594) and the ‘numerus Hnaudifridi’ (RIB 1576). The latter was probably an irregular unit named after its commander with the Germanic name Hnaudifridus (Bosanquet 1922). The term ‘Germanus’ is therefore once again used in the context of inter-regional communication networks with other Germanic groups, and on this occasion was inscribed upon the threshold of a doorway which physically operated as its right of passage.
The etymological origins of the deities worshipped at this shrine – Mars Thingsus, Beda, Fimmilena, Baudihille and Friagabis – have received a great deal of discussion for their overt usage of Germanic phonetic traits. As factions from the Frisian and Tuihantian tribes, located on the east-bank of the Rhine, it seems likely that they were familiar with a Germanic language. The term ‘Thingsus’ has been convincingly related to a Germanic word for ‘assembly’, which is attested in later Germanic languages such as Old High German ding, Icelandic thing, Old Frisian thing, and Old Norse thing (see Hübner 1885; Stephens 1885; Simek 1993: 203; Green 1998: 34). ‘Beda’ and ‘Fimmilena’ are even enshrined in later medieval law codes recorded in Frisia, whereby ‘fimelthing’ and ‘bodthing’ are juristic procedures within such assemblies (see Hübner 1885; Simek 1993, 203; Green 1998, 34). This suggests that the ‘temple’ itself is likely to have been perceived and used as an assembly by these groups. The deities venerated by the ‘numerus Hnaudifridi’ also betray Germanic philological traits, but accentuate the more martial connotations of the ‘assembly’. Baudihille is thought to relate to Germanic *badu meaning ‘battle’, and *hild ‘war’ (Bosanquet 1922). Friagabis is cognate with *frî meaning freedom, and -gabis ‘to give’, which is cognate with Old High German gabe and Gothic gabi (Bosanquet 1922; Green 1998, 39; Clay 2007). This type of ‘assembly’ is not only of the greatest significance to these Germani at Housesteads, however, but to those Germanic tribes of a much earlier period. The afore-mentioned assembly convened by the Ubii at Cologne in honour of the ‘German people’ and the ‘German name’ had been similarly in honour of ‘Mars’ as well as other ‘common gods’ (Tacitus Histories 4. 63). We find a striking parallel here in terms of the context in which the Germani are attested in both these literary and epigraphic sources of data. Whilst the etymological origins of Housesteads’ Germanic personal names and deities have attracted this wealth of academic attention, little interest has been paid to the term ‘Germanus’ itself, which is hardly surprising since it is dismissed as a Roman label of minimal interest. As already mentioned, the meaning of the term is supposed to lie in philological obscurity. However, if one were to translate the roots of this word in the same manner as the other deities and personal names recorded at this site, one finds that the second stem -man relates to Proto Germanic *mann meaning ‘man’ (Bradley 1908: 99). This is noted in the names of other Germanic tribes of suspected indigenous origin such as the ‘Marcomanni’ and ‘Alammani’ (Wolfram 1997, 40; Wells 1999, 190). The first stem Ger-, however, compares with a West Germanic term for spear, which is attested as Old English gar, Old Saxon ger, Old High German ger (Bosworth and Toller 1898: 361–2). It is also found in Germanic personal names recorded in the Roman period such as the Batavian Maloger (CIL 3. 3577) and Garisianus (CIL 13. 7833). This is consonant with early Medieval Germanic uses such as in the names Luidger and Gerwulf (Searle 1897: 257), and suggests that Germanic speakers of Roman date had already adopted the onomastic and etymological function of this stem. Tacitus (Germania 6) intriguingly comments on the fact that the spear was the most important weapon by which the Germans were identified, but he only knew the native word of ‘Frameae’ for this weapon, not possessing much lexicological knowledge of their native tongue. That the ‘Germani’ may well have meant something along the lines of ‘spear men’ to groups such as the cives Tuihanti, however, is further corroborated by references to the ‘Gairethinx’ ‘spear assembly’ in the Langobardic Law codes (Fruscione 2005: 21). This relationship between spears and assemblies can be even extrapolated backwards to the Roman period on the basis of the well-known passage, which states that at such concilia young warriors were allocated a shield and spear in order to mark their initiation into manhood: ‘then in the assembly one of the chiefs or his father or his relatives equip the young man with shield and spear: this corresponds with them to the toga, and is the youth’s first public distinction’(Germania 13).
Figure 11. Plan of the excavated temple dedicated to Mars Thingsus, Housesteads. Birley, E. 1962. Archaeologia Aeliana (fourth series) 40: figure 2 page 122.
Figure 12. Reconstruction of the entranceway leading into the temple dedicated to Mars Thingsus. The altar served as a lintel supporting the circular arch, but the other altar/lintel from this entrance has not been recovered. Author’s own adapted from Corpus Signorum Imperii Romani. Corpus of Sculpture of the Roman World. Great Britain, vol. 1, fascicule 6: numbers 157 and 161. The British Academy.
The historical and epigraphic sources of evidence would therefore point to the fact that a ‘German’ represented an adult male, who having passed adolescence was eligible to participate within such assemblies. One should also take into account the fact that Mars Thingsus is depicted with both these weapons on the doorway erected in the Germani’s honour (Fig. 12). As Nico Roymans (2004, 248) has recently pointed out, tribal assemblies in the Rhineland were often utilised to raise troops for service within the Roman army. The site at Housesteads perhaps offers us another explicit example of this, or at least the prominence of such assemblies within the military units which were raised. The ideological importance of the shield, and particularly the spear, to many men serving in the Roman auxilia is also manifest in the common funerary motif depicting the soldier armed with these weapons. This motif made a significant revival in the Roman parts of the Rhineland in connection with native auxiliary soldiers (Carroll 2001: 121; Roymans 2004, 230). It also travelled to wherever the recruits from these regions were dispatched (Figs 9 and 13), but it was not simply restricted to Germanic tribes. In some respects, a ‘German’ may have figured as a particular type of male with martial obligations and duties, which came to be represented through both literate and iconographic means from the very beginning of the groups’ appropriation of the epigraphic record. The archaeological data casts further light on the life-span of this particular form of identity, as is evidenced by remnants of spears and shields recovered from water sanctuaries on the Lower Rhine, which have been interpreted as votive deposits offered by veteran soldiers upon their completion of service (Nicolay 2001). A review of such sources of material might suggest that the ‘German’ status would have been part of many men’s life-cycle. However, it was not intended to be a permanent state, unless one died as a soldier – as was the case with those Germans at Rome and in Britain. One wonders whether what differentiated ‘Crotilo Germanus’ from his brother ‘Vindicianus’ was the latter’s completion of military service?
Figure 13. Gravestone from Croatia dedicated to the memory of the Batavian Imerix. Roymans, N. 2004. Ethnic Identity and Imperial Power: the Batavians in the Early Roman Empire: figure 10. 2 page 228.
Summary and Discussion Previous attitudes towards the Germani in Roman studies have fundamentally underestimated the significance of the term in providing insights into the self-perceptions of Germanic tribes incorporated within the vicinity of the Rhine frontier. This paper only partially compensates for such neglect and leaves a great many aspects still unresolved; but its intention has been to highlight the discrepancy with the prevailing orthodoxy and to underline the potential utilisation of often overlooked terms recorded in Britain and Rome. Archaeological attempts to assign an identity to the Germani have been consciously avoided since the end of the Second World War (see Todd 2004: 9–10; Wolfram 1997, 10), but this has been at the expense of leaving unexplained both literary and epigraphic references to the self-perceptions of peoples so designated. This proposed understanding of the term conforms with much of what is already espoused about Germanic tribes situated on the Rhineland, but it allows the societies themselves a more active participation in the construction of their own identity.
The provisional model put forward here suggests that the term ‘Germanus’ came to be used in reference to a certain number of groups located on the Lower Rhine. The conflict between east–bank migrants and indigenous ‘Gallic’ groups may have contributed to a heightened ethnic signalling amongst the Germani. However, the etymological significance of the term may have, in any case, leant itself to a generic use. The importance of assemblies within these tribes might have meant that each had its own quota of Germani. Participation in native government was presumably socially restricted (see Roymans 1990: 30), but it may have been accessible to greater numbers in times of political activity, no doubt coinciding with the period when more warriors were required.
By the mid-first century A.D., the exploitation of native assemblies to mobilise young men into Roman service significantly increased the numbers of Germani on the Rhine frontier. Whilst the term may have derived via Germanic-speaking groups, this does not necessarily suggest that it was restricted to them since, through the context of the army, other Germanic words were adopted into the languages spoken on the Rhineland. The protracted usage of the term ‘Germani’ to refer to many societies protecting the Rhineland may, however, explain its suitability for the provincial titles of Germania Inferior and Superior. The most overt expression of a common German identity amongst these people is documented during the Batavian Revolt, but here it can be seen as a culmination of mutual dissatisfaction related to military matters without any suspicion of nascent nationalism.
To the Roman government, people from the provinces of Germania Inferior and Superior would be loosely referred to as the ‘Germani’ but the term may also have had a restricted sense to the groups themselves, which has only partially been investigated here. Such significance would undoubtedly have shifted according to different temporal and regional settings, since it is the prerogative of each generation within every society to negotiate the identities associated with the names it lives by. In Britain the term ‘Germanus’ seems to have been associated with military units of the Batavi, Cuberni, Frisiavones, Frisii, Tuihanti and Vangiones, who had been initially permitted into Roman empire in order to participate in military defence. The term is occasionally given to persons of suspected social and economic importance within militarised contexts but exact martial positions are never provided. A more detailed assessment of how Roman literary sources refer to individual ‘Germans’ in military situations may throw some light on the matter. If certain groups did perceive themselves as Germans, however, then this type of identity may have been acted out in particular forms of behaviour, some of which leaving a material trace. Thus, it is hoped that future research about both the emic and etic references to these people in ancient sources, as well as the signification of ethnic relationships in the archaeological record, might be able to develop further the identity of the Germani in Roman studies.
Department of Archaeology, University of Sheffield
Acknowledgements This paper stems from a presentation given in Jennifer Baird’s and Judith Rosten’s session entitled ‘Developing Identity in Roman Studies’, which I also delivered for the Archaeology Department at the Universiteit van Amsterdam for those otherwise unable to attend the TRAC conference. I would very much like to thank all those who contributed in the discussions afterwards for their stimulating, as well as challenging input, which facilitated the refinement of this paper. I am also greatly indebted to the Worldwide Universities Network (WUN) whose three-month exchange programme with the Universiteit van Utrecht enabled me to gain a much broader perspective of various theoretical issues, which have been fed directly into this paper. Additional obligations must also be given to Carol van Driel-Murray, lecturer at Universiteit van Amsterdam and Lauran Toorians, guest lecturer at Universiteit van Utrecht, for both granting me access to their own articles prior to publication. I would also like to thank Maureen Carroll Reader in Roman Archaeology at the University of Sheffield as well as Professor Nico Roymans at the Vrije Universiteit Amsterdam for kindly allowing me to make use of illustrations from their own publications. Lorraine Seymour PhD candidate at the University of Sheffield must also be accredited for much needed assistance with digital analysis. Finally, a special word of gratitude must be directed towards my PhD supervisors John Moreland and Philip Shaw, again at Sheffield University, as well as to Martin Henig, Professor at UCL, for reading through drafts of this paper and making innumerable recommendations for improvement.