An Island Nation: Re-Reading Tacitus' "Agricola"

by Katherine Clarke
An Island Nation: Re-Reading Tacitus' "Agricola"
Katherine Clarke
The Journal of Roman Studies
Start Page: 
End Page: 
Select license: 
Select License



There's a sense of place about islands -even when the! 're semi-detached -that sets them
apart from continents.'

hanc oram novissimi maris tunc primum Romana classis circumvecta insulam esse
Britanniam adfirmavit.
(At that time a Roman fleet circumnavigated this most remote shore for the first time and
confirmed that Britain ~vas an island.) (Tac.,Agric. 10.4)


Tacitus' Agricola is one of the most tantalizingly enigmatic of ancient texts. Coming from the pen of one who was to become a renowned historian, it is notoriously hard to place in generic terms. It fails to conform to any commonly accepted model of political history, and yet, as I shall argue, it has much to tell us about Tacitus' views of Roman political life. We can turn to the parallel of the Germania for another possible way out of the dilemma, and yet the ethnographic details which the Agricola undoubtedly encompasses could hardly be seen as its main focus. The most natural cast to give the work draws on its ostensibly biographical aspect. Commemorating the res gestae of Tacitus' father-in-law, Agricola, is the purpose signalled to the reader from the first sentence onwards: 'to hand on to future generations the deeds and values of distinguished men' ('clarorum virorum facta moresque posteris tradere'). All of these interpretations have had their proponents. But I shall argue here for a different reading of the Agricola, one which not only highlights an aspect of the text which has tended to be sidelined, but also provides an interpretative framework within which some of the other, more extensively treated, themes may be reconsidered. My reading of the Agricola is focused not on the state of Rome under the emperor Domitian, nor on the customs of the inhabitants of Britain, nor even on the figure of Agricola himself, but on the actual location of his res gestae. I shall consider how Tacitus' portrayal of Britain itself may ultimately offer us insights into Agricola, Domitian, and Roman political life.

Two possible approaches to the geographical aspect of the Agricola have already been explored and exploited. Dorey saw the question of location as very much subordinate, almost incidental, to Tacitus' relation of Agricola's achievements: 'the account of the achievements of Agricola's predecessors as governors of Britain and the description of the country itself and its peoples merely serve as a prologue and a backcloth to the great drama in which Agricola is to be the protagonist.'2 This view of the location as a mere stage and of its description as a piece of literary scene-setting is one against which I shall argue strongly. I shall propound a reading of the Agricola in which 'the description of the country' is indeed the ideal location for the res gestae of Agricola, but one which permeates the work and offers a crucial key to its wider interpretation. In contrast to Dorey's 'backcloth', a focus on location might seem to promise a traditionally empirical approach to the Agricola, which entails an attempt to map Agricola's exploits and travels onto modern Britain. Archaeological research has greatly enhanced the rewards for those who pursue this approach, and, if \ve take the

* This article has been developed from a paper given readers for their many illuminating comments. A11 at the Laurence Conference on Roman Britain, held remaining errors and infelicities are, of course, my in Cambridge in May 1997 and organized by Mary own. Beard and John Henderson. I should like to thank the I hT.Crane, TWODegrees Ft'est. A Mhlk along Eng- organizers and the participants in the conference for land's Jleridian (1999), 350. their many helpful suggestions and criticisms. I T. A. Dorey, "'Agricola" and "Germania"', in should also like to thank Chris Burnand, Fergus T.A. Dorey (ed.), Tacitlrs (1969),4. Millar, Chris Pelling, and the anonymous YRS

G \Vorld copyr~ght: The Soc~ety for the Promotion of Roman Studles 2001


text at face value and trust in its credentials as an account of Roman incursions into the British Isles and, in particular, of the campaigns of Agricola, then its value for reconstructing the history of that period is considerable.

I have chosen, however, to explore a different angle on the issue of geographical location: namely, the light shed by the Agricola on the Roman view of Britain as a geographical entity and as a potential subject nation. I aim to show that Tacitus manipulates traditional aspects of this geographical conception, in particular Britain's place in relation to the Continent, in ways which illuminate not only Roman Britain, but also the nature of Rome it~elf.~

A consideration of the Agricola, which takes the geographical vision of Britain as its starting-point, may unexpectedly offer new insights into set pieces such as the pre-battle speeches of Agricola and Calgacus, as well as bringing into sharper focus the question of how Romanness could be constructed and negotiated. The key to this reading lies in a feature of Britain which has remained contentious to this day, periodically threatened and fiercely defended, namely its insularity.


In order to appreciate the significance of Tacitus' portrayal of Britain as a geographical entity, it is important to consider the tradition into which he inserted his account. A first impression of the history of geographical thought and the expansion of knowledge through conquest and ex~loration tends to focus attention on the Mediter- ranean \Grid.\!Lo indeed would be;nterested in doing more than creeping around the shores of the Mediterranean, listing cities, villages, and peoples along the way in the manner of Ps-Scylax or Ps-Scymnu~?~

It is possible to appreciate the allure of the eastern land-mass of Asia, much of which mould be imagined, constructed, and reconstructed firstly by the companions of Alexander the Great, and then by succeeding generations. But the cold, damp, miserable North-\Vest held considerably less appeal. Tacitus himself gives some idea of the re~ellent nature of this Dart of the world in the Germania: '~hosi who wanted to change {heir habitat in the travelled by sea rather than by land, and the vast Ocean that lies beyond and acts, so to speak, as a barrier, is rarely approached by ships from our world. In any case, quite apart from the danger of an inhos~itable and unknown sea. who would leave behind Asia or Africa or Italv to make for Germany, ugly in its landscape, harsh in its climate, and grim to live in and to look at, unless it were his own land?' (2.1). In what terms could such liminal areas be represented?'

I shall argue that it is precisely the remoteness of the North-West and particularly of Britain itself which Tacitus exploits in the Agricola; that Britain's isolation, and specifically its insularity, underpins Tacitus' exploration of other themes within the work. There was indeed a long-standing tradition about the nature of Britain. According to Tacitus. many authors had treated the island and its inhabitants before (10.1).He mentions tmo of ihese authors by name: Livy and Fabius Rusticus, who are de'scribed as 'the most eloquent of ancient and recent authors, respectively' ('Livius veterum, Fabius Rusticus recentium eloquentissimi auctores', 10.3). In spite of these earlier treatments, Tacitus still has something to contribute, because the island mas only fully conquered at the time of Agricola ('tum primum perdomita est', 10.I), thus offering new opportunities for speculation to be replaced by knowledge. Furthermore, I shall argue that the conquest of Britain, as related in Tacitus' Agricola, not only enhanced empirical

E. O'Gorman, 'hTo place like Rome: identity and For these periplus texts, see C. Lliiller (ed.), difference in the Germania of Tacitus', Ramus 22 Geographici Graeci .Winores I (18j j). (1993)~135-54, sees the Germania as an 'exploration j For discussion of the text of the Germania, its of a country (Germany) in search of the ideological relation to other ethnographic works, and later treat- (Roman) self' (135). Many of her ideas about the ments of the land and people, see the excellent Germania are also relevant to discussion of the commentary by J. B. Rives, Tacitus. Germania (1999). Agricola.

knowledge of the island, but actually altered its location on the mental map of the Romans.

As Ogilvie and Richmond note in their commentary, Caesar, Strabo, Pomponius Mela, undoubtedly Posidonius, and, long before them, Pytheas of Massilia had compiled accounts of Britain.6 But the conceptual framework for Tacitus' Agricola may be unexpectedly enhanced by first considering the accounts of those writers who dealt not with Britain, but with the JVest coast of Africa. The periplus texts attributed to Hanno, king of Carthage, and Scylax of Caryanda may appear at first to have little to offer our understanding of the Agricola, but I shall argue that they give us a productive insight into the geographical conceptions which characterize the world beyond the Pillars of Hercules, a world to which Tacitus' Britain must at least partially belong.

JVe know of Hanno's voyage from the Greek translation of a Punic inscription, supposedly set up in Carthage to commemorate the expedition.7 Hanno apparently set out from Carthage in c. 480 B.C. to explore the JVest coast of Africa and to found Liby- Phoenician cities. It has been suggested that the voyage referred to in the text never actually took place, and that the text represents rather an attempt to construct non- Greek alterite', but for the present purpose of assessing the conceptual geography of the world outside the Mediterranean, questions of fictionality need not concern us.'

The text purports to recount a voyage down the coast into increasingly hostile and alien territory and its inhabitants. The world of the Outer Ocean is one in ~vhich the landscape, its inhabitants, and their languages become threateningly different from those of the Mediterranean. All of these features will be important elements in understanding the use which Tacitus makes in the Agricola of the earlier geographical tradition. Hanno and his men come across lands of fire and incense from which streams of fire flowed into the sea, making approach impossible ($15-17). Not far into the journey they encounter elephants ($4)and wild beasts ($7).From then on the division between men and beasts becomes blurred. There are Troglodyte men who run more swiftly than horses ($7), and lvild men clad in animal skins ($9). This confusion culminates in the discovery of the gorilla women, about whom I shall say more shortly. The language of the inhabitants matches the increasing oddity of the landscape. Early in their voyage the Carthaginians take on board interpreters, but on reaching the land of the Ethiopians, they discover that their language cannot be understood by even the Lixite interpreters ($ I I).

Furthermore, the world beyond the Pillars is a world of islands. Hesiod and the Homeric epics had established that the western Ocean contained the Isles of the Ble~sed.~

Hanno's voyage confirms the insular nature of this region. He passes the island of Cerne soon after leaving the Pillars of Hercules ($8).I shall return to the significance of Cerne below. But it is also an island. and furthermore an island within a lake within another island, which provides the location for the most startling of the experiences to befall Hanno and his companions, and which marks the end of the journey. 'On the lake [sc. mithin the island] was another island full of wild people. By far the majority of them were women with hairy bodies. The interpreters called them gorillas. JYhen we chased them we mere unable to catch the men for they all fled from our hands since they climbed well and defended themselves \\ ith stones. We captured three women, however, who bit and scratched those who led them and did not want to follow. So we killed them and flayed them and took the skins to Carthage. For we sailed on no further, for our food

See R. LI. Ogilvie and I. .I.Richmond (eds), De See Homer, Od. 4.j63ff. on Elysium in the Ocean Vita Agricolae (1967)~16 j. (see Strabo 1.1.j for the islands of the blessed in the 'Text, translation, and commentary are provided context of Homeric geography); Hesiod, Works and by J. Ramin, Le Periple d'Hannon. The Periplus of Days 170-7 describes these islands in some detail. Hanno (1976). Pindar, 01. 2.68ff. continues the tradition. Of course

For the suggestion that the voyage is an imagined the delightful nature of these islands is in contrast rather than a real one, see C. Jacob, Geographie et with the inhospitable places encountered and ethnographic en Grece ancienne (1991), 73-84. It is, of described by the explorers, but the abnormality of the course, also possible, even likely, that not only the island landscape and inhabitants remains a constant voyage, but also its epigraphic commemoration was theme. fictional.

was running out' ($17). And here the account ends. An island on an island, together with its strange inhabitants, marks the limit of exploration.

The unfamiliar nature of what lay beyond the Pillars is confirmed by the mid- fourth-century B.C. text attributed to Scylax of Caryanda." After performing the traditional circuit of the Mediterranean basin and returning to the Pillars of Hercules with a neat sense of closure, the author then breaks out beyond the confines of the Mediterranean and enters the unfamiliar territory explored by Hanno. After passing some of the landmarks noted bv Hanno. he reaches the island of Cerne and runs into problems. 'What lies beyond Ceine canndt be reached by ship because of the shallowness of the sea and the mud and seaweed' (r4~ Kkpvq~ 6~ vfioou rh 67'tkKEtva O~KETIBoti xhorh 6th ppa~htqtrx Bah&r~q< nqh6v ~rxi +OKO<)($1I 2). This is an island \\ orld where the sea

~ai is no longer navigable and the normal rules of physics do not apply.

Around a century after the purported voyage of Hanno down the coast of Africa, Pytheas of Massilia sailed north on emerging from the Pillars of Hercules. His remard was the reputation of being utterly untrustworthy; his name became synonymous mith tall stories. Strabo picked up on Polybius' misgivings concerning the evidence produced by Pytheas on Britain and its environs. At least Euhemerus, the archetype for geographical invention, claimed to have sailed to only one country, Panchaea, fictitious though that may have been. But Pytheas claimed to have explored the whole northern region of Europe as far as the ends of the earth (pk~pt tGv TOO ~oopou nspktov), an assertion which no one would believe, not even had Hermes himself made it (Strabo 2.4.2).

The claims which Pytheas apparently made concerning the area are strongly reminiscent of the morld constructed by the authors of the voyages of Hanno and Scylax, a morld of islands, strange landscapes, and strange peoples. According to Strabo, Pytheas claimed the existence of an island called Thule, six days' sail north of Britain and near the frozen sea (1.4.2). Pytheas managed to convince the great Hellenistic scientist, Eratosthenes, that the island existed, but, according to Strabo, men mho had seen Britain and Ierne (Ireland) did not mention Thule, although they did speak of other small islands around Britain (1.4.3). 'Pytheas of Massilia tells us that the areas around Thule, the most northerly of the Britannic islands, are the most remote . . . but, from the other writers. I can find nothing. out -neither that there exists an island called


Thule at all, nor that the areas right up to the point \\here the summer tropic becomes the arctic circle are habitable. I believe that the limit of the inhabited world on the northern side lies much further south than this. For modern mriters have nothing to say about any country beyond Ierne (Ireland), which lies close to the north of Britain and is the home of men \\ ho are utterly wild and lead a miserable existence because of the cold; and therefore, in my opinion, the northern limit of our inhabited world is to be placed there' (2.5.8).

The sea around the Thule area, as described by Pytheas, bears a striking resemblance to what Scylax found around Cerne. 'He [sc. Pytheas] talked about those regions in mhich there was no longer either land in its own right, or sea, or air, but a kind of substance made up of all these elements, resembling a "sea-lung", in mhich the earth. the sea. and all the elements are held in sus~ension: and this is a sort of bond to hold them together, but you can neither walk nor sail on it' (pfi~e xopeutov pfi~e xho~ov) (Strabo 2.4.1). Just as Scylax described, the islands beyond the Pillars of Hercules seem again to belong to a world in which elements behave strangely, making places inaccessible and quite literally hard to define, as land and sea are merged into one.

The world beyond the Pillars may seem to bear the hallmarks of a fictional creation. As I have already noted, Strabo found Pytheas' account incredible; Jacob has suggested that Hanno's voyage may never have taken place, except in the imagination, and indeed the text relating his adventures may be a later fake. The similarity in sea-quality betmeen that around Cerne and that around Thule begins to hint at some profound geographical problems. The use of 'stock literary places' in geographical works has been the subject

lo For an extensive and detailed treatment of Scylax, portolano del Mediterraneo, Biblioteca di Studi see A. Peretti, I1 Periplo di Scilace. Studio sul primo Xntichi 23 (ed. E. Gabba) (1979).

98 KATHERINE CLARKE of some recent discussion, in particular concerning the islands of Cerne and Thule, whose precise locations prove hard to define. It has been suggested from the fact that the name 'Cerne' was applied to several different places -opposite the Persian Gulf according to Ephorus (Pliny, ,YTH 4.35) and beyond the Pillars of Hercules according to Eratosthenes (Strabo I .3.2)-that the term 'non rappresenta una frontiera geografica, ma un confino fantastico'." The island of Thule has been found similarly elusive because 'nella letteratura antica il nome Tule indica l'estremita settentrionale dell'ecu- mene'.12 So, Cerne as the indicator of western limits, and Thule of the north may not be representative of 'real' islands at all, but part of a fictional and schematic mapping out of the world.13 If this is the world in which Tacitus' Agricola takes place, then any use of the text for the tracing of Roman campaigns in Britain may be compromised. Accurate mapping and pure fiction are not polar opposites, but the way in which real locations are perceived and 'mapped' is very much a matter of ideological and, as I would argue in the case of Tacitus' Britain, intertextual con~truction.'~ Before focusing on Tacitus, I make one final allusion to the Greek geographical tradition and to its construction of the non-Mediterranean world. Homer, according to Strabo the father of geography, propagated the idea of the encircling Ocean. The encircling Ocean was important because it was suggestive of a limit, and in spite of the best efforts of the Hellenistic geographers such as Eratosthenes to argue that more lay beyond, many, including Strabo himself, clung to the Homeric world-view.15 For those propounding a 'Roman' vision of the world there were clear advantages to be found in a geography which specified an outer boundary, which could be presented as the natural limit to imperialism.16 Such notions may be brought into play when considering the imperialist exploits of Agricola and their presentation by Tacitus. Holyever, it is not immediately clear holy the idea of the Ocean as the outer limit of the Empire can be reconciled with the fact that the Ocean itself, as seen in the exploratory tradition, is a place of islands, and furthermore islands lvhich may be seen as potential conquests. \!here are we to locate Britain -within or lvithout the boundary that conveniently delimits Roman imperial aspirations?17 I shall discuss in the following sections the ambiguous location of Britain, in the ill-defined and nebulous lvorld that characterizes the Ocean beyond the Pillars, but none the less the appropriate setting for Roman imperial res gestae. In particular, I shall consider Tacitus' manipulation of different geographical traditions in ways which both highlight and exploit this ambiguity.

In many respects the world of Tacitus' Agricola resembles that of the Ocean beyond the Pillars, found by the Greek explorers to be full of islands and physically abnormal. Dorey says of the Germania, 'there is a feeling of remoteness from the subject','' clearly referring to Tacitus' lack of personal interest, but in a purely spatial sense both Germany and Britain, regions of the North-\Vest, are portrayed as being remote from Rome, in a liminal position on the edge of the known world. There is no doubt but that Britain is

" G. Amiotti, 'Cerne: "ultima terra"', CISA 13 l6 That Roman commanders from Pompey onwards, (1987) 43-9: 'it represents not a geographical bound- and partly in emulation of Alexander the Great, made ary, but a fantastical limit.' political mileage out of 'reaching the Ocean' is appar- l2 F. Cordano, La Geographia degli antichi (1992), ent. See, for example, Cicero, de Imperio 33; Sallust,

107: 'in ancient literature, the name Thule indicates Letter of Mithvidates 17: Plutarch, Pompey 38,~-3: the northern extremity of the inhabited world.' Plutarch, Caesav 58.6-7, For Alexander, see Pompe-

13 For the close relationship between geographical ius Trogus, Histoviae Philippicae 12.7.4. literature and fiction, see J. S. Romm, The Edges of l7 D. Braund, Ruling Roman Britain: Kings, Queens, the Eavth in Ancient Thought (1992). Got'ernors and Emperovs from Yulius Caesar to Agricola

l4 The idea of 'constructed landscapes' is interes- (1996), 12, expresses this neatly: 'Britain lay both in tingly developed in relation to the Gevmania by Ocean and beyond Ocean, so that the conquest of O'Gorman, op. cit. (n. 31, especially at 136. Britain was also the conquest of Ocean itself.'

'5 See Strabo 1.4.6 for Eratosthenes' belief that the l8 Dorey, op. cit. (n. z), 17. \I-estern Ocean formed not a limit but a route to the world beyond and ultimately to India.

ambiguously placed. The northern shore is faced by no other land (10) and is beaten by the great and open sea (vastum et apertum mare), although Britain's place here within the encircling Ocean is worth noting. The sea around Thule in the Agricola precisely recalls that of Pytheas' account and also Scylax' tale concerning Cerne. It is 'sluggish and heavy' (pigrum et grave), qualities mhich are said to make rowing difficult. The sensible explanation has been provided by Burn: 'the meaning of this has been elucidated only by men who have been there. The Romans would have encountered the North Atlantic Drift Current running against them.'19 Ogilvie and Richmond too offer this as an explanation for what Tacitus describes. All three scholars also mention Pytheas, but to disappointing effect. For Burn, Tacitus' description is too geographically specific to be an allusion to the world of Pytheas; for Ogilvie and Richmond, 'Pytheas was certainly alluding to the freezing sea round Iceland, but Tacitus describes a different phenom- enon', thus denying the point of contact.20 Their commentary may well be correct in asserting that Pytheas' Thule was Iceland and that of Tacitus Shetland. But this very confusion signifies that Tacitus' portrayal of these regions is coined in the currency of moveable islands and jelly-like seas, the remote morld beyond the Pillars, making use of standard literary topoi for geographically distant areas.21

The ambiguity between land and sea is yet another feature of the island world which appears in the Agricola. Tacitus rejects a detailed discussion of the Ocean, but he does mention the supremacy of the sea in so far as it cuts its way deep inland around Britain, blurring the natural boundaries (10). Not only this, but the overthrow of the natural order is manifested in the confusion between day and night. A feature of the most remote part of Britain, and indeed a function of that remoteness, is the existence of night which is paradoxically 'bright' and 'short' (clara and breeis: 12). The implications for the question of definition and ambiguity are made explicit: 'you could hardly distinguish between evening and morning twilight' ('finem atque initium lucis exiguo discrimine internoscas'). The sun can be seen all night, not rising and setting, but simply gliding across (transire). It must be, says Tacitus intriguingly, that the flat extremities of the earth, mith their low shadow, do not project the darkness. Whatever the theory behind this, there is a clear connection between extremity, remoteness, and the failure of the natural morld to behave according to the usual rules. Britain's failure to adhere to boundaries is further exemplified by the leadership of a woman, Boudicca, 'for they make no distinction between the sexes in their commands' (16).

The view of Britain as a remote island, lying on the edge of the earth, is, however, under threat in the text. One of the quotes with which I started concerns the moment in history, namely the time of Agricola's campaign, mhen the island nature of Britain was firmly established, but I shall argue that this recognition also marks the first stage in Britain's removal from the island world of the Ocean. It has been argued that the passage in mhich this discovery is stated forms part of a digression, thus sidelining it from the main narrative.22 Ogilvie and Richmond compare the digressions in the monographs of Sallust, which, they claim, are similarly used to separate the principal stages in the narrative (p. 164). However, it has more recently been suggested that the

l9 A. R. Burn, 'Tacitus on Britain', in T. A. Dorey for example, to Himilco's exploration of the outer (ed.), Tacitus (1969), 40. See also A. R. Burn, 'Mare parts of Europe, in which were found seas that were pigrum et grave', Classical Reoiezc 63 (1949)~ 94, for difficult to navigate. Certain phrases in Avienus' this passage as an accurate description of a 'phenom- account of Himilco's voyage are particularly reveal- enon familiar to sailors of small craft in the Pentland ing. See, Ora Maritima 121: aeqzloris pigri and 128: Firth'. naoigia lente et languida repentia.

20 Ogilvie and Richmond, op. cit. (n. 6), 173. 22 Even such sensitive treatments as that of W. Lie

2' Further weight is given to this connection by the beschuetz, 'The theme of liberty in the Agyicola of observation by J. 0.Thompson, History of Ancient Tacitus', Classical Quarterly 16 (1966), 135-6, can Geography (1948), 148-9, that there was an extensive find little to link the description of Britain with the tradition in geographical writings of discovering and surrounding narrative. describing sluggish seas in remote areas. He refers,

digressions in Sallust are anything but digressive.23 I would argue that for Tacitus too the digression on the geography and ethnography of Britain is integral to the work. Britain's insularity is not just a backdrop, but an element which must be continually reassessed and redefined. The other quote which I introduced at the start provides the cue for the forthcoming discussion of Britain as an island, which, though remote, may be seen as also 'semi-detached'.

This is the point at which to note that sailing out through the Pillars of Hercules was not the only means of reaching Britain. An alternative route, and one which would give Britain a quite different identity from that of the Oceanic world, was across land through Gaul. This was the route taken by Caesar; it was also the picture given by Strabo. who wresented his descri~tion of Britain as a continuation of that of Gaul. and stressed the English Channel as a medium for communication rather than as a barrier (Strabo 4.5.1-5). Britain was the natural next step in Rome's conquest of the world, 'une suite logique de celle de la Gaule', according to Di~n.~~

This, I shall argue, is the other main confi~uration of the location of Britain which Tacitus and the characters of the Agricola reveal and exploit in the course of the text.

Crucial to the process of redefinition is the intellectual conquest by the Romans of their prospective new territory. This intellectual conquest must go alongside the literal one effected in the campaigns for Tacitus to be able to claim that Britain was first completely conquered (perdomita) at the time of Agricola. The Romans cannot start to reassess the place of Britain in their world-view until they have become informed about the island.25 So we hear that the Orkney islands were discovered and subdued, but had been unknown (ignotae) up until that time (10);finding out about the earliest inhabitants is difficult even now (parurn cornperturn). But Agricola's mission is to reveal, uncover, bring all within the compass of Roman knowledge. In his third season of campaigning, he uncovers new peoples ('novas gentes aperuit', 22); the fifth season brings still further discoveries -the subjugation of peoples unknown until that time ('ignotas ad id tempus gentes', 24). But Agricola employs his knowledge discriminately: 'he knen- everything, but he did not follow it all through' ('omnia scire, non omnia sequi', 19).

The final show-down at >Ions Graupius provides the context not only for the military clash, but also for the battle over knowledge of Britain. Calgacus, the Scottish chieftain and Agricola's worthy opponent, spurs on his men n-ith reference to the remoteness and seclusion (longiquitas ac secreturn) which have so far kept his land out of the Roman grasp. He characterizes the Romans as utterly unfamiliar with all that surrounds them: 'fearful through ignorance, looking around at the sky itself and the sea and the woods, all unknown entities' ('trepidos ignorantia, caelum ipsum ac mare et silvas, ignota omnia circumspectantes', 32). Agricola totally reverses this picture in his speech of encouragement: 'Britain has been discovered and subdued' ('inventa Britannia et subacta', 33). Although the Romans may not have a precise knowledge of the region, these are not new peoples and an unknown battle-line ('novae gentes atque ignota acies', 34). From the British perspective, as portrayed by Tacitus, the less that is known about

23 See T. Wiedemann, 'Sallust's Jugurtha: concord, Ogilvie and Richmond entitle their commentary on discord, and the digressions', Greece and Rome 40 chs 10-12 'The Ethnography of Britain'. Xly aim is in (1993)~ 48-57. K. Wellesley, 'Tacitus as a military part to reinstate the Britanniae situs as being of at least

historian', in T. A. Dorey (ed.), Tacitus (1969), 69, equal importance to thepopuli. places the geographical 'digression' of the Agricola in 24 R. Dion, Aspectes politiques de la geographie anci- the category of origines et situs, which Tacitus dis- enne (19771, 254. cusses at Ann. 4.32ff. as being an integral part of the 25 Burn, op. cit. (n. 19), 37, notes the disproportion- historiography of imperialism. The description of ate geographical interest in the work devoted to Britain would thus be motivated by the fact that the islands. 'The Romans seem to have had an appetite island was then for the first time fully conquered. for British island names', adducing also Pomponius This argument is somewhat supported by Tacitus' Xlela's mention of 30 Orcades, Thule, and 7 announcement that he is moving on to discuss 'Britan- Aemodae. niae situm populosque'. It is disappointing that


Britain by the Romans the better; for the Romans, intellectual conquest is going to be an important part of the take-over.

Tacitus' image for the Roman intellectual victory over Britain, when he states that Britain is the largest of the islands which Roman knowledge embraces ('quas Romana notitia complectitur', 10)has serious implications for Britain's insular nature. Agricola's campaigns have the effect of altering Britain's status as an island, previously free from the Continent.

In his account of the history of British incursions, Tacitus recalls the Britons' view of their insularity. The Germans were disadvantaged in having only a river, rather than the Ocean, as a defence against the Romans: 'they were defended by a river, not the Ocean' ('et flumine, non Oceano defendi', 15). Before Agricola's invasion, the Homeric encircling Ocean could be seen as a defence betn een the Continent and the British Isles. Britain lay far out in the Ocean itself. The previously remote, Oceanic nature of Britain is reinforced by Tacitus, who comments that opposite the North shore was no further land; the shore was battered by the vast, open sea (10). Calgacus may use the Oceanic nature of Britain as a spur to his men at Mons Graupius -'no land lies beyond us' (ultra). But the advent of Agricola changes the conceptual geography of this part of the world, in ways which even Calgacus cannot deny.

Instead of the Ocean acting as a barrier between Britain and the Continent, it now acts as the medium by which the Romans can embrace Britain within their grasp. As Tacitus says, it was at the time of Agricola that Britain was first circumnavigated by the Roman fleet. The paradox is that although this proved that Britain was an island, and therefore remote and theoretically beyond the scope of Rome's imperial aspirations, the act of circumnavigation was a crucial stage in the removal of Britain from the Oceanic world beyond the Pillars, and its gradual attachment, both administratively and in the Roman mental map, to the world of the C~ntinent.~~

The circumnavigation led to the embrace (comnplectitur) of Britain by Roman knowledge, but also to a change in its insular status; a geographical realignment, so that it now fell within the compass of the Roman world, not in the world of the elusive western islands. The resultant exclusion of Britain from the world of the Outer Ocean is reinforced by the note that Agricola's men who circumnavigated Britain could not do the same for Thule. This island was sighted, but no more, allowing it to remain firmly outside the Roman world.

As the campaigns move further northwards, the island status of Britain is continually reassessed. The fourth year brings the Romans to a possible terminus, the narrow strip of land between the Forth and the Clyde. If the glory of Rome had not forbidden a halt, says Tacitus, a limit would have been found in Britain itself ('inventus in ipsa Britannia terminus'). The garrisoning of this strip of land would have resulted in the enemy being cut off, as if onto another island ('summotis velut in aliam insulam hostibus'). The island mentality is at the fore; islands mean separation, isolation, and defence. But here, the idea is to move the boundary of Britain to a point which the Romans can reach. What lies beyond would join the quite separate world of Oceanic islands, a world which is not of immediate concern to Rome -an extreme example of

26 For earlier formulations of the idea that the Ocean that was scarcely believable. In the context of neb- might encircle and encompass Roman pourer, note the ulous island-urorlds, the phrase vfioov hrrtoroupivqv ambitions of Pompey (Plut., Pomp. 38.2-3) and fino payieovq (Plut., Caes. 23) is particularly striking. Caesar (Plut., Caes. 58.6-7) The flexible definition of Here, there is no suggestion that Caesar's conquest where precisely the Ocean ran, beyond Britain or would threaten the insular nature of Britain; rather, before one reached it from Gaul, could, of course, be its insularity is proof of Caesar's daring. As Pelling useful in justifying either Roman expansion in that observes, Pompey's allusion to Caesar 'calling the direction or the exclusion of the island from imperial pools of uncertain depth an Ocean' ('Oceanumque aspirations. I owe to Chris Pelling the point that vocans incerti stagna profundi', Lucan, Civil Way elsewhere Plutarch describes Caesar's British expedi- 2.571) might suggest that Caesar himself made much tion as taking him beyond the confines of the Empire, of the Oceanic adventure in his dispatches concerning that is beyond the Oceanic barrier and into a world the British expedition.

the way in n-hich the Romans could redraw their mental map to match their aspirations and achievement^.^'

This option is rejected. Britain will be conquered wholesale. The following year's campaign opens up a new issue, the status of Ireland, and raises interesting possibilities for the shifting alignment of Britain as either an island or an adjunct to the Continent. Strabo had made it clear that Ireland fell outside the intended limits of the Roman Empire, being utterly barbarian and undesirable, a place of savagery, cannibalism, and incest (4.5.4) Tacitus' Agricola at least harbours ambitions in that direction. As for Britain, the insular nature of Ireland is firmly established, but it is interesting that the extreme remoteness of Ireland found in Strabo's account is tempered by Tacitus. The island is smaller than Britain, but larger than those in the Mediterranean (nostrum mare), contexts of comparison which immediately draw Ireland and the norld of the Ocean towards the L'lediterranean Roman norld. The notion expressed in ch. 24, that conquest of Ireland mould help to link together the western parts of the imperiurn -Britain, Ireland, and Spain -brings these Oceanic lands into the embrace of Rome again, redrawing the map of the Greek geographical tradition, which had placed them outside the world of the hlediterranean periplus. The nen- geographical grip on Ireland, which its conquest mould bring, is foreshadowed by the inclusion of Ireland in the intellectual grasp of the Romans already. Tacitus says that the harbours on the island are fairly n-ell- known (cogniti) through the reports of traders. '4s I discussed abo\re with regard to Tacitus' Britain, knowing about a place is a half-way stage to conquering it with arms.

Before the arrival of Agricola, the sea, which was so much a part of the British landscape as to penetrate even far inland, proved a bonus to the British. The sixth year of the campaigns mas to change all that. At this point, Agricola set in motion a marine counterpart to the landborne offensive: 'The n-ar mas pushed forward simultaneously by land and sea: infantry, cavalry, and marines mould meet up in the camp and compare the conquest of land and the enemy with that of the Ocean' ('hinc terra et hostis, hinc victus Oceanus', 25). According to the tales of capti\res, the Britons were amazed, as though 5th the recesses of their own sea opened up, the last refuge for the defeated had been shut off'. The Britons consider the Ocean to be tlzeir sea (szri maris), and now it is being taken over by the Romans, who already have the l'lediterranean as their own (nostrum mare). The confidence brought by the fact that the Ocean itself, the very definer of Britain's former identity, is now in his power, leads Agricola to abandon once and for all the notion of a terrninus within Britain, and to seek the real ternzinus, the Oceanic ternzinzds which encompasses the whole island, and ultimately the whole earth. There is no need now for subdivisions which would leave a newly created 'island' still to be conquered.

By the time of the battle of L'Ions Graupius, there can be no doubt in anyone's mind that the insular status of Britain has been eroded; its position among the nebulous islands that lie beyond the Pillars, the haunts of Hanno, Scylax, and Pytheas, has been taken away by Rome's conquest. The conquest is both intellectual, drawing the whole island into the realm of Roman knowledge, and forcing a redrawing of the mental map of the area, and real, in the sense that Rome has actually reached the terminus of Britain, paradoxically using as the medium of travel the Ocean itself. This part of the Outer Ocean, like the llediterranean, had non- in a sense been appropriated by the Romans as nostrunz mare; it no longer belongs to the Britons.

27 The mental dix-ision of Britain into a 'Roman' and from conquest, see Strabo 4.5.3. Braund, op. cit. a 'non-Roman' section is clearly seen in Appian's (n. 17), 149, however, interestingly points to the way preface, in which he outlines Roman acquisitions, in which Scotland, under the name of Caledonia, was including 'the larger and better portion of it [sc. used by Flavian poets to express the rulers' aspirations Britain], since they did not want the remainder at all' to conquer Britain, telling against the idea of a (App.,RomanHistory.Pref. 5). For the unprofitability northern segment of the island which could be men- and consequent undesirability of Britain as an element tally detached and dismissed. in Roman rhetoric concerning the island's elusiveness AN ISLAND NATION: RE-READING TACITUS' AGRICOLA 1°3

The speeches before the battle of Mons Graupius provide the dramatic opportunity for this gradual process of redefinition to be brought into clearer focus.28 Calgacus starts in encouraging fashion. Now is the dawn of freedom for the whole of Britain (libertas toti Britanniae). You are all (universi) gathered here. free from servitude. But it is

2 u

apparent that all of these terms have now taken on a more sinister aspect. Libertas is a theme to which I shall return. But the united nature of Britain was a feature of its life as an island. distinct and self-contained. In the face of earlier Roman incursions. Boudicca had taken up the war for all (universi, 16);it was natural for Calgacus to make the same appeal. But we have seen the way in which the intervening campaigns of Agricola have now made the union of Britain desirable in Roman eyes. The option of dividing Britain has been weighed up and rejected. The Romans, having encircled Britain geographically, are now happy for it to be united as a bundle which can be easily adjoined to the Empire.

No land lies beyond us, says Calgacus; not even the sea is free from the threat of Rome. But we have already seen that Ireland, lying beyond Britain, is part of the Roman plan for future conquest, and one which would take away the prospect of libertas from Britain. When Calgacus goes on to stress the isolation of Britain from subject peoples 'we have eyes which are undefiled by contact with domination' (30) -he is referring to the Gauls. Britain was, for him, the safe island retreat from the Continent; for the Romans the concern is that Britain should not be able to see liberty on the other side, out into the Ocean. But I have argued that the Roman redefinition of Britain relies on looking towards the English Channel, not the Ocean. Yet again Calgacus is made to speak in terms which curiously echo the wishes of the Romans. Rather than looking outwards, he is made to focus his sights on precisely the link that Rome wants to effect.

Up until this point, says Calgacus, Britain was defended by its very remoteness and seclusion (recessus ibse ac sinus); now the terminus of Britain is revealed. The effects of ~~ricola's'cam~ai~is

in terms of bringing Britain within the realm of Roman knowledge are accepted. As Calgacus' speech proceeds, the picture of Britain it puts forward seems to be increasingly in line with that presented in the narrative itself. Agricola himself needs only to summarize the same picture in his speech, claiming to have secured the limit of Britain @is Britanniae). Returning to the terminology of Tacitus' 'digression' in chs 10-12, it could hardly be made more explicit that the conquest is as much one over the landscape (situs), the insular nature of Britain and its geographical position, as over the inhabitants bobuli) themselves.

One might see [his nek identity of Britain as being one of 'semi-detachment', or perhaps more accurately as one of 'semi-attachment', since the process is in the direction of assimilation rather than disjunction. As a result of Agricola's campaigns the island undergoes a redefinition which results in a loss of strict insularity, an erosion of the remoteness which was crucial to its independence from Rome and was manifested in its similarity to other landmarks of the physically strange world beyond the Pillars. Kow Britain is made to adhere to the Continent, to minimize the real and conceptual gulf of the Channel, which had previously provided an important divider and kept Britain as a feature of the Oceanic world. The possibility that Britain might be subdivided and the northern half allowed to go free, redefined as an island in its own right, has been rejected. With the Roman grasp embracing not only the Channel and the whole of Britain, but even the sea beyond, the realignment is complete. However, as Crane's comment on semi-detached islands notes, even these retain a 'sense of place that sets them apart from continents'. And, as I shall argue in the next section, it is the continuing ambiguity in Britain's geographical status which provides a fitting context for Tacitus'

28 They may also be, according to R. Martin, Tacitus battle speechgiven by the British chieftain, Caratacus. (1989), 43, used as a means of establishing that the Xlany of the themes are very similar to those raised by Ag~icolais to be seen as a historical work. R. .\I. Calgacus -libertas, airtus, maiores, the family. Just Ogilvie, 'An interim report on Tacitus' "Agricola"', as in the case of Calgacus, Caratacus is defeated in ANRW I1 33.3, 1720, also sees in the pair of speeches spite of his brave words. His main speech in oratio in the Agricola an echo of those attributed to Hannibal recta comes after his defeat, when he has been taken and Scipio in Livy 21.40-4. For a Tacitean parallel, to Rome. see Annals 12.34, where Tacitus alludes to a pre-

exploration of the behaviour and values of the protagonists, both conquered and conquerors.


It is one of the many complexities of the text that the same picture of Britain, as an island whose insularity is under threat and which is encircled by Roman power and knowledge, can be used as an encouragement to both sides. It is not hard to see how Agricola might use such a picture to give confidence to his troops, but the reasoning of Calgacus is more difficult to understand. For this, we need to consider what else Calgacus feeds into his picture. The definitions of Britain may to a large degree coincide, but Calgacus then turns his attention to the portrayal of the Romans, and herein reveals another significant shift from the traditional picture. I shall argue in this section that the ambiguity of Britain as a location for Roman res gestae is reflected in the difficulty in defining both its inhabitants and its conquerors.

It was standard practice in geographical and ethnographical works to portray peoples on the edge of the earth as pirates and brigands. Not only those far from the centre of power, but all those characterized by their 'out-of-the-way-ness' (k~~ontopoq) are prone to piratical behaviour. And one of Rome's greatest achievements was the suppression of such piracy. Throughout the first century B.C. this n-ould be a major route to fame and power for politicians such as Lucullus and Pompey, and n-e know that it was an image still being propagated by ~ugustus.~~

But various leaders and authors had tried to turn this picture around. Sallust's Letter of Mithridates (22) designates the Romans as latrones gentium. Calgacus, in the words of Tacitus, follows in this tradition of criticism. The Romans are, he says, the brigands of the world (raptores orbis). Not only this, but they have not been satisfied with East and West (non Orzens, non Occidetzs satiaverzt). The implication is that they also n-ant the North, the region of Thule in the schematic layout of the earth. But we knon from authors such as Vitruvius and Strabo that the Romans belonged at the centre of the earth. It n-as their location at this climatologically privileged point that made them naturally prone to hegemony.30 Now their imperial ambitions had led them to the very edges of the earth, and they had started to behave according to the stereotype of the barbarians whose natural habitat was on these perimeters. This is one possible reading and I am arguing in general in this paper for the importance of place and location, but I shall return to consider whether this picture of geographical determinism, whereby the Romans behave piratically because of their new location in barbarian lands, stands up to close scrutiny.

Furthermore, the progress of the text reveals that these 'Romans' are not necessarily Roman at all. The army is made up of Gauls, Germans, and even some Britons. The Romans do not just behave like barbarians; they actually are barbarians. Just as this makes it impossible to draw a sharp distinction between the troops of Calgacus and Agricola, so too does it reveal a lack of unity within Agricola's army itself. And the same tension between unity and diversity among the Britons is also revealed in the work.31

29 For Lucullus, see Plut., Luc. 23; for Pompey, see See Vitruvius 6. I. 10-1 I;Strabo 17.2. I, but especi-

Cicero, de imperio Gnaei Pompeii 34-5, in which the ally also 6.4. I. anti-pirate command is seen as the spring-board for 31 It is worth remembering that Agricola himself is a the granting to Pompey of the command over the war Gaul, a detail which further confounds any attempt at against hlithridates through the Lex hlanilia. For strict divisions. On the implications of this, see I. A. Augustus, see his claim at Res Gestae 25 ('mare pacavi Richmond, 'Gnaeus Iulius Agricola', JRS 34 (1944), a praedonibus'). This may indeed refer to the war 44. Braund, op. cit. (n. 17), 155, comments that against Sextus Pompeius in Sicily in 39-6 B.c., but Agricola's up-bringing in Gaul keeps him away from the point is made more general in Horace, Odes4.5.19: the corruption of Rome itself, a point to which we 'pacatum volitant per mare navitae' and in Suetonius, shall return. Life of Augustus 98, in which the Alexandrian sailors hail Augustus: 'per illum se vivere, per illum navigare'.


Just as Agricola's army may be affected by their new location on the edges of the earth, behaving piratically as befits their 'out-of-the-way-ness' (?~7ontopoq), so too are the Britons affected by the presence of Agricola's men -the processes of acculturation at work. Tacitus famously catalogues in ch. 21 the effects of Roman influence on the way of life of the inhabitants of Britain. 'To induce a people, hitherto scattered, uncivilized and therefore prone to fight, to grow pleasurably inured to peace and ease, Agricola gave private encouragement and official assistance to the building of temples, public squares, and private houses. He praised the keen and scolded the idle, and competition to gain honour from him was as effective as compulsion. Furthermore, he trained the sons of the chiefs in the liberal arts and expressed a preference for British natural ability over the trained skill of the Gauls. The result was that in place of rejection of the Latin language there came a passion to command it. In the same way, our national dress came into favour and the toga was everywhere to be seen. And so the Britons were gradually led on to the amenities that make vice wleasant -arcades. baths. and sumwtuous banquets. It was called civilization by those to whom it was all new, when really it was a feature of their enslavement.' I shall return to the implications of the last sentence, but for now simply note the change for Britain and the lifestyle of its inhabitants brought about bv the-advent of Agricola.

coking back to ~argacus, it is striking that he is seen to possess and display the most Roman of virtues, virtus itself. He is introduced as a man outstanding in virtus and nobility (29). He attributes this to the Britons under his leadership: they have virtus ac ferocia (virtue and fierce courage) (31). Virtus is what the Romans lack. 'Do you think that their virtus in war matches up to their wantonness in peace?' asks Calgacus (32). Besides this, he attributes familial piety to the Britons: Agricola's army is without any familia, no wives to spur them on, no parents to chide if they run away from battle, and his whole speech ends with an appeal to the ancestors and future generations (maiores and posteri) -a standard appeal for Romans to make.

Even more strikingly, Calgacus' entire speech is a masterpiece of Roman oratory: full of sententiae ('omne ignotum pro magnifico est'); rhetorical questions; balanced antithesis ('non Oriens. non Occidens satiaverit' -where even the sentiment ex~ressed entails notions of balance, spatial in this case). There is a grand rhetorical crescendo: 'auferre trucidare rapere falsis nominibus imperium atque ubi solitudinem faciunt pacem appellant' ('robbery, murder and rape they falsely call empire: they create desolation and call it peace', 30). Here the rhetorical practice of definitio, defining one's terms, is shown off.32

We thus have the paradoxical scenario in which Calgacus, the barbarian chieftain, is more skilled in the art of speaking Latin than are the Romans themselves. We know that his definitions are more accurate than theirs: he knows better how to use their language. The Romans are barbarian brigands whose Latin is decidedly suspect: Calgacus and his men are the repositories of Roman virtus and proper Latin usage. Has the process of acculturation resulted here in a wholesale exchange of identities?

The question of language usage seems to me key. In the famous passage on Romanization at ch. 21, Tacitus describes the acquisition of Roman habits and the accoutrements of civilization: 'idque apud imperitos humanitas vocabatur, cum pars servitutis esset' ('it was called civilization by those to whom it was all new, when really it was a feature of their enslavement'). Tacitus makes clear that the Britons have not adopted the correct term for what is' happening. This cannot be due to their short acquaintance with the language, since Calgacus, far away in the North of Scotland, does not share this problem. Furthermore, he is actually able to correct the Latin used by the Romans themselves: his understanding of the terminology of imperialism is more accurate than theirs. It cannot be the case that one's Latin improves through contact with the invading Romans; otherwise, those in the South of Britain, rapidly becoming Romanized, would be far more advanced than Calgacus. It is, of course, in any case a conceit that Calgacus was a fluent and accomplished Latin orator.

32 See Cicero, De Oratore I .42(189) for the definition features which particularly characterize the thing
of definitio as 'a short piece of description of those which we wish to define'.

Rather than being improved by contact with the Romans, it appears that the Britons of the South have been corrupted. Their adoption of the twisted Roman terminology justifying conquest is one example of this. They have learned to speak as the Romans do, and that means in a corrupt way. By contrast, Calgacus, hidden away in the most remote part of Scotland, on what could even be thought of as the second island, has not yet been corrupted. His Latin is pristine. As he claims for all Britons: 'we are uncorrupted and unconquered' (integri et indomiti), an extravagant claim and not a true one. The unity of Britain, to which Calgacus appealed and which would have suited the Romans too, is a fa~ade. Some have been taken over and brought into the Roman mind- set, but there still remains a pocket which has not succumbed. The difficulty then is to consider what the state of uncorrupted Britain is, and what Calgacus stands for.

I have mentioned the many ways in which Calgacus seems to epitomize Romanness. He is almost more Roman than the Romans themselves. So we cannot say that Agricola's Roman army is corrupting a barbarian nation, but rather that there may be two types of Romanness in play. Calgacus is representative of the fact that Old Rome is to be found in the most remote parts of the Empire, or even beyond the Empire's bounds.33 Old Roman virtues and grand Latin speeches are located at the edge of the earth, in the most peculiar world of the Oceanic islands, and as far from Rome itself as one can imagine.


This is one way of understanding the prominence of Calgacus, as the representative of Roman virtus and Roman eloquentia. But the text is introduced as a commemoration of the illustrious achievements of Agricola. How can his opponent be allowed to eclipse the hero of the work to the extent that he and his land are revealed as the repository for the kinds of values we would naturally associate with the clari viri of Rome?34 I return now to the question of Britain as the ideal location for Agricola's res gestae and to some explanation of why the issue of Britain's geographical status is much more than a mere backdrop to Agricola's campaigns.

I have set out the way in which Tacitus may be playing on two geographical conceptions of Britain: placing the notion of Britain as part of the remote island world of the Outer Ocean, together with Cerne and Thule, alongside the alternative conception of Britain as an adjunct to the Continent, the logical follow-on from Gaul. Agricola, as a native of Forum Iulii (4), and educated in hlarseilles, belongs by birth to the second of these two worlds, that of Gaul, the Continent, and the Roman Empire. His military apprenticeship under Suetonius Paulinus introduced him to Britain and, as Tacitus says, he got to know his province well (noscere proeinciam) (5). His final qualification as the ideal candidate for the governorship of Britain came with his magnificent term in command of Aquitania (9). Public opinion demanded that he should be appointed to Britain, not because he himself suggested it, but because he was the right man (par eidebatur). Agricola, with experience of both Gaul and Britain, was the ideal person to effect the transformation of the latter into an adjunct to the Continent, to make it 'semi- detached'.

But Britain was also the ideal location for Agricola by virtue of the fact that its insularity was still in question; its geographical alignment remained ambiguous. Returning to my initial quotation, I suggest that Britain was the perfect location for

33 O'Gorman, op. cit. (n. 3), 147-9, discusses this acknowledgement of the fact that times hare changed theme in relation to the Cermania,raising the question since the Republic and a new type of heroism is called of whether Germany could be seen as equivalent to for. However, I disagree with his view that Calgacus' primitive Rome, or 'a reenactment of early Roman portrayal as worthy opponent is designed primarily to history'. In the Germania,just as in the Agricola, this increase the stature of .Agricola (44). It seems to me question finds no easy solution. rather that Calgacus actually embodies what Agricola 34 See R. %lastin, Taritus (2nd edn, repr. 1994), himself might ha\-e been like if he had not been a 39-49. He discusses the work as one in Lvhich the Roman general at the time of Domitian, and subject

figure of Xgricola as hero is maintained partly through to the concomitant constraints.

Tacitus' commemoration of Agricola's exploits precisely because it was still significantly set apart from the Continent, even though in the process of becoming semi-attached.

I have argued that Tacitus presents a picture of Britain as the repository of Old Roman values. In particular, Britain in its pristine state, as represented by Calgacus at Mons Graupius, is the location of true eloquentia and virt~s.~'We know from the Annals and the Dialogus de Oratoribus something of Tacitus' views on the opportunity for eloquentia and virtus at Rome under the Principate. Oratory was dead; virtus rewarded with a rapid fall. That we may be encouraged to read these events in Roman Britain in the light of Rome itself is surely suggested by the framing of the work. Tacitus starts the A~ricolawith a lament over the hostilitv shown at Rome at the time of Domitian to


eirtus: 'so hostile to virtue are the times' ('tam . . . infesta virtutibus tempera', I). He moves on to mention examples of men who made a protest at the lack of senatorial freedom of speech (eloquentia) under Nero and Vespasian -Thrasea Paetus and Helvidius Priscus. Their eulogists, Arulenus Rusticus and Herennius Senecio, are recalled. as is the fact that not onlv was eloauentia in the Senate disallowed. but so too was eloquentia in the form of books recording the suppression of senatorial eloquentia. The books were burned in an attempt to obliterate the voice of the Roman people (vox populi Romani) and the libertas of the Senate. Rome of old explored the limits of libertas; by the time of Domitian it was exploring the depths of ser~itus.~~

The Agricola ends, as has been noted by various scholars, with a passage strongly reminiscent of Cicero's consolatio in De Oratore I11 for the death of L. Licinius Crassus, the most renowned Roman orator until the time of Cicero him~elf.~' It is not clear to what extent Agricola might be thought to display the ideals of libertas embodied by Crassus, and to what degree he, in Domitianic Rome, can be only a pale imitation of this Republican figure.

Tacitus was, of course, not alone in observing the moral decline of Rome. For Sallust virtus had been destroyed by the removal of the external enemy of Carthage in the second centurv B.C. and it is interesting that Tacitus follows Sallust at least in the


monograph form he chose to give to this work. Martin, among others, has observed the strong evocation at the start of the Agricola of Sallust and Cato, perhaps signifying the similarity of theme or sentiment between this work and the writings which emerged from and lamented the dying days of the Republic.38 But Britain. which belonged at least in Dart to the Oceanic world of islands and is - defined in the ~~ricola It is by its remoteness, was a place of eloquentia and liberta~.~~ specifically in the the North of Scotland, to which the Romans considered giving special island status, where these most Roman of values are on display. Britain, as an island nation, forms the antithesis of Rome in Agricola's day, as-described by Tacitus. Is Britain the paradigm for Rome of the past, and particularly of the Republic? The start of the Dialogus, in which oratory and eloquence are seen as features of the Republic, and destroyed by the Principate, might offer support to this reading. It seems likely from the process described in Agricola 21,by which contact with Romans corrupts the perceptions and modes of expression of the Britons, that Britain of the future may become like Rome of Domitian's day. But the corollary does not necessarily hold. It is not in any sense clear that Rome of the past was like Britain at the time of Agricola, 35 Again, Caratacus provides an illuminating parallel. 93-113. The classic treatment is C. Wirszubski, His post-defeat speech is given in Rome, not in Libertas as a Political Idea at Rome during the Late Britain, and it draws him even closer to 'Old Rome' Republic and Early Principate (1950). than even Calgacus can come, since Caratacus not 37 See Martin, op. cit. (n. 34), 48; R. M. Ogilvie, 'An only is eloquent, but directly echoes the pre-death interim report on Tacitus' "hgricola"', ANRW I1 speech of Cremutius Cordus, a historian condemned 33.3, 1718. for his libertas under Tiberius. Caratacus' claim at 3s Martin, op. cit. (n. 34), 41; M. M. Sage, 'The 12.37 that execution, the final suppression of his treatment in Tacitus of Roman Republican history libertas, mill bring him eternal memorial is strikingly and antiquarian matters', ANRW I1 33.5, 3385-419, reminiscent of Cordus at Ann. 4.35. This parallel is esp. 3388. all the more resonant given the framing of the Agricola 39 M. A. Giua, 'Paesaggio, natura, ambiente come with references to loss of senatorial libertas at Rome. elementi strutturali nella storiografia di Tacito', 36 The bibliography on Tacitus' views of libertas is ANRW I1 33.4,2897, argues that the Agricola reveals vast, and a discussion of the subject lies far outside that 'la lontananza e l'isolamento rispetto a1 cuore the scope of this paper. See, for example, M. Ham- della civilita siano considerati garanzia di autentica mond, 'Res olim dissociabiles: principatus ac libertas', liberti'. Harcard Studies zn Classical Philology 67 (1963), except in its opportunities for eloquentia and libertas. We cannot explain the relationship between Britain and Rome and the values propagated in each place simply in terms of a time-lag. The question of location must be reintroduced. In what sense can Britain be seen as an appropriate venue for the res gestae of Agricola? Beyond the very obvious answer that it was in Britain that Agricola did indeed carry out his relatively successful campaigns, I suggest that Tacitus turns the location into more than just the real and suitable backdrop for those achievements; that he makes the question of geographical identity central to the commemoration of Agricola's life. Agricola came from a family of eloquentia and virtus. His father, Iulius Graecinus, was renowned for his eloquentia and sapientia and was punished by the emperor Gaius for these virtutes (4). His mother, Iulia Procilla, was a model of chastity (rarae castitatis). Agricola himself was something of a philosopher; he also enjoyed marital harmony we may note Calgacus' appeal to the familial piety of his troops. Agricola was a man of virtus, so thought Domitian at any rate, for on hearing about Agricola's successes, his reaction was that 'other talents could be just about ignored, but the virtus of a general was the preserve of the princeps' (39). Agricola's danger from Domitian lies in the fact that Domitian is hostile to virtus, in which Agricola excels. Agricola is finally driven headlong to destruction by his own virtues (suis virtutibus) and by the faults of others (42). Thus Britain, as the location of traditional Roman virtues, provides an appropriate setting for the playing out of the yes gestae of Agricola, a hero of the old style.40 Here there mav be some instructive ~arallels to draw from Pellinp's work on Germanicus as a u ~e~ublic'an-style hero.41 It is iAteresting that both Agricola and Germanicus enjoy their greatest achievements right at the edges of the Roman world. The context of Germanicus' final episode in Germany before his recall is strikingly similar to that of Agricola's campaigns in Britain: a world of open sea (aperta Oceani), islands (insulae), and hidden shoals (occulta vada) (Ann. 2.21). Indeed. some of Germanicus' shi~wrecked men are carried to ~ritain, which the;;elate as a place of oddity and ambigui&, in terms both of its geography and its inhabitants @nn. 2.2~+).~~ The landscapes in which these imperial commanders operate are evocative of the heroic Republican world and the historiography which went alongside its aggressive imperialism.43 Tacitus would, of course, famously link such geographical descriptions with the relating of Republican achievements at Ann. 4.32, and lament their passing, as Giua points out. That observation might shed light on why even the harsh, heroic landscape of Britain and the North-West cannot ~rovide the ideal location for the ex~loits of Germanicus and Agricola. Their opportunities at Rome are limited by imperial jealousy. But even here in the remoteness of Britain the atmosphere of libertas, virtus, and eloquentia does not provide the perfect environment for Agricola's glorious res gestae. Instead, like Germanicus, he remains curiously ~nfulfilled.~~ Agricola's key-word is not excellence, but moderation (moderatio). He constantly weighs the situation up, and is not allowed to complete his campaign. One problem in assessing the behaviour of Agricola and of his potential subjects is the on-going ambiguity over what it means to be Roman, the difficulty over identification which I discussed in the ~revious section and to which I must return. For Braund. Calgacus and his views are belied by the narratil~e, leaving Agricola for the most part See Sage, op. cit. (n. 38), 3388-93, for the framing not one at all. So, the traditional historiographical of the Agvicola in a context of Republican style virtus. topos of the commander surpassing a mighty bound- The opening chapters make clear that a significant ary may be toned-down in accordance with the times, contrast is to be drawn bet~veen the possibilities for and the new geographical image of the encompassed outstanding deeds and their recording in the past and Britain may be appropriate for the nexv muted hero- in the present. ism. I owe this point to Chris Pelling. 41 See C. B. R. Pelling, 'Tacitus and Germanicus', in 44 For imperial incidia against Agricola, see Ag. 39; A. J. Woodman and T. J. Luce (eds), Tacitus and the for that against Germanicus, Ann. 2.22 (metu incid- Tacitean Tvadition (1993), 59-8 5. iae);2.26. I. Shatzman, 'Tacitean rumours', Latomus 42 On the landscape of Germanicus' campaigns, see 33 (1974), 574, suggests a further parallel in that Giua, op. cit. (n. 39), 2887-90. rumours surrounding both Agricola and Germanicus, 43 However, it is worth noting the shift in the image unusually for Tacitus, almost al~vays cast them in a of Ocean as no longer the proof of a great campaigning more favourable light than reality suggests. exploit (as for Caesar), but now a muted barrier, or without criticism. He dissociates Agricola from Roman corruption, stressing his Gallic origins and claiming that Agricola's civilizing mission is misinterpreted and manipulated by the Britons in ch. 21, who 'adopted more Romanization and a different Romanization from that which Agricola had offered'.45 But, if Agricola is not fully Roman, then neither are the inhabitants of the land which I have characterized as one of Romanitas, eloquentia, and virtus. Calgacus and his men are more than just worthy opponents of Agricola. They fall into a historiographical category of adversaries who are 'more Roman than the Romans'; so Roman that they traditionally bring out the best in their Roman counterparts. But in the Domitianic world of the Agricola, where achievement must be muted, the usual topoi misfire and Agricola is not inspired by Calgacus to reach the heights of success. Furthermore, the notion that Calgacus knows Latin better than the Romans themselves is valid only if we define Romanitas and eloquentia in the out-dated terms that seem to suit heroes like Germanicus and Agricola himself. There may be yet a further twist in this argument, in so far as speaking Latin well, perhaps always and certainly in Rome of this period, means being disingenuous, artful, rhetorical, even misleading, and speaking the corrupt Latin adopted by the southern Britons. If this is so, then it is no longer clear that Calgacus really is the better speaker of Latin, the better preserver of Romanitas. VII. EPISODES IN THE OCEAN: FICTION AND REALISM The Agricola is full of paradox and ambiguity. It is remarkable that we might even contemplate that Romanitas in the age of Domitian might be found in the world of permanent day and jelly-like seas. It is significant that Agricola's campaigns in this most remote corner of the world known to the Romans should ~rovide the context for Tacitus' exploration of 'how to be a good man under a bad emperor'.46 If it is necessary to go so far in order to maintain one's integrity, and even then in a compromised way, then the implications for Rome itself must be bleak. It is also startling that the reader might be left with some reservations about the very subject of commemoration. Tacitus manipulates the geographical tradition to create a shifting vision of Britain. He requires Britain to have a remote identity, both in order that he might make a point about the location of Romanitas and, more specifically, for Agricola's strengths to find a suitable setting. But Agricola himself threatens to undermine Britain's insularity, to remove it conceptually from the world beyond the Pillars and turn it into an adjunct to the Continent. What can there be to commemorate in the life and career of Agricola, if he is the force behind the destruction of this pristine island world which encapsulates the values that he himself holds dear? The precarious and ambiguous status of Britain is nowhere more clearly apparent than in the pair of episodes which frame the great show-down at AIons Graupius, in which, I have argued, the process of Britain's geographical redefinition is brought into closest focus. The first episode concerns the mysterious Usipi, who murdered their Roman commanders, hijacked three ships, and set off on a voyage around the whole of Britain. Coming from Germany, they belong to the region which had once provided the context for the achievements of Germanicus, like Agricola, thwarted in his pursuit of e~cellence.~'The Usipi are in many ways like Agricola's troops, comprising a cohort which had been levied in Germany and transferred to Britain. They come closest to fulfilling Calgacus' belief, as expressed in his speech before the battle of Mons Graupius, that the barbarians who made up Agricola's army would recognize their roots and switch 45 Braund, op. cit. (n. 17), 63. it provide the subject for Tacitus' other work in 46 For Tacitus' belief in the opportunity for virtue monograph form, but it acts as a foil in all kinds of and heroism even in a wicked age, see J. Percival, ways. Germanicus' exploits are notoriously unful- 'Tacitus and the Principate', GMR 27 (1y80), 127-9. filled (Ann. 2.26 for the recall of Germanicus on the Another key example is Lepidus in Annals 4.20. cusp of victory); Domitian's own expedition \\-as 47 The importance of Germany for our understand- incomplete (Ag. 39). ing of the Ag~icolais a recurrent theme. Not only does sides. Indeed in ch. 32, he uses the Usipi as the exhortatory exempla. \re hear of no such process in the aftermath of the speech; rather Agricola's troops are the men who have been corrupting Britain as they advance. But the Usipi had found in Britain an invigoration of their 'edge of the earth' existence. Their behaviour is the least civilized to find a place in the work. Having killed the centurion and soldiers who had been set over them to teach discipline, they set sail like pirates; indeed they were taken for pirates ('pro praedonibus habiti', 28) and some ended up being sold as slaves. Their difficulty in procuring food led to acts of cannibalism (28) by contrast with the relatively normal diet of the inhabitants of the island (12). Their level of barbarism was so great that they appeared uncivilized even in the context of the wild North-\Vest, with Britain providing the ultimate opportunity for behavioural libertas. But the Usipi, antithetical though their behaviour may be to the civilizing force of Agricola, are also paradoxically reminiscent of some of the qualities associated with him. Their murder of the Roman soldiers might have been a dreadful crime, but it won them fame, or at least notoriety ('magnum ac memorabile facinus', 28 ).48 Fame and commemoration motivate the writing of the Agricola, a point to which I shall return, although Agricola himself is forced to be moderate in his seeking of such renown. In this respect the Usipi, in a perverse way, are allowed to succeed where Agricola fails, although there is of course a considerable difference between acquiring fame for great deeds and notoriety for heinous crimes.49 Furthermore, the circumnavigation of Britain by the Usipi forms a parallel to that accomplished for the first time by the Romans with Agricola, in the passage quoted at the start. The Romans were able to establish that Britain was an island, 'since it had been sailed around' (circumvecta, 10). As I have argued, this voyage marks an important stage in the reconceptualization of Britain; the confirmation of its insularitv is the first stage in the threatening. of that status. But for the Usipi, the circumnavigation of Britain (circumvecti Britanniam) is part of their liberation from the constraints of civilized behaviour and from their subjection to Roman commands. Their voyage is hardly unproblematic, culminating as it does in the loss of the crucial component, the ships. It is noteworthy that this loss occurs through ignorance of the necessary skills ('amissis per inscitiam regendi navibus', 28), in stark contrast to the knowledge and preparation associated with Agricola and his army. Nevertheless, the Usipi with their achievement of fame and their exploitation of Britain as a place of invigoration, the recapturing of a libertas which had now been lost in their native Germany, give us a final glimpse into the vision of Britain as a fully-fledged island, free from the corruption brought by continental attachment, and it is important that their visit should involve a reaffirmation of this insularity through circ~mnavi~ation.~~ There follow the speeches, fighting, and aftermath of the battle at Mons Graupius, on which I have commented above, and in the course of which the new vision of Britain as semi-detached, an adjunct to the Continent, is adopted. It is interesting that the final triumphal act instigated by Agricola after the battle should involve still further use of the Ocean as a means of asserting control. Agricola orders the commander of the fleet to sail around Britain (circumcehi Britanniam), using the terror of Rome as his defence in the face of trouble. Unlike the voyage of the Usipi, this one is well-planned, and ultimately successful. Progress is assisted by favourable winds isecunda tempestate) and the reputation of the Romans ifama) (38). Thus two Roman voyages around Britain frame that undertaken by the Usipi, and the Romans emerge as the masters of Britain's waters and hence of its insularity We should perhaps in any case recall Calgacus' be dangerous is very relevant to the question posed by designation of the Romans as raptores orbis (30.4). the Agricola of how to be a good man under a bad 49 However, the fact that there could be a fairly thin emperor. line to be drawn between great deeds and transgress- 50 Note the contrast drawn earlier between the Ger- ive ones was neatly encapsulated in the tales sur-mans, for whom the only defence from Rome Lvas a rounding Alexander the Great, whose conquests and river, and the Britons, who could hide behind an ambitions could be seen as both mighty and hybristic, Oceanic barrier. Here, the Usipi benefit from the although this still does not match the transgressions geographical advantages afforded by Britain, and of the Usipi. However, the point that greatness could denied to their own people. But there is a further Oceanic episode which provides an alternative way of configuring the text. The report of events at Mons Graupius elicited a jealous response from Domitian. We are implicitly reminded of why the Usipi were fortunate by comparison with Roman generals such as Germanicus and Agricola, in so far as their eaperness for fame and adventure could be realized. whereas Agricola must hold back. Popular reaction demanded that Domitian show restraint in his reining in of Agricola, and the province of Syria was to be offered as a sop. But, predictably, the opportunity for Agricola to excel was snatched away. A freedman was dispatched to offer the province to Agricola, but only if he was still in Britain. The story went that the messenger met Agricola's ship in the Channel (in ipso freto Oceani), and returned to Domitian without even addressing Agricola himself. So, whereas the Oceanic episode concerning the Usipi and preceding Mons Graupius symbolized opportunities for exhilaration, that which follows the crucial turning point in Britain's Oceanic status reveals yet another lost opportunity for Agricola. His potential for glory is unfulfilled, but he himself had been responsible to some degree for the development of a new geographical conception of Britain, in which the Ocean becomes a medium for suppression, rather than a symbol of libertas. The fact that the whole British narrative in the A~ricola (chs 10-40)8 , falls within the " framework of the Ocean, neatly mirroring its real physical location, is suggestive of the importance that might be attached to Britain's geographical status, and in particular to its insularity. Our first extensive encounter with the location in ch. 10 stresses the Oceanic nature of the place and indeed hints at the possibility of more detailed study of the Ocean itself, a possibility which is rejected as exceeding the scope of the present work and, in any case, already treated by other writers." However, Tacitus does acknowledge that there is nowhere more dominated by the sea than Britain ('nusquam latius dominari mare'); in the expanse of water lies the key to Britain's identity. Our very last glimpse of Britain in the work is the near-encounter with Agricola in the Channel; again the insularity of Britain is at the fore, although by now the Ocean has been reduced to a mere strait Cfretum). But there is something unsatisfactory about the Ocean and the episodes which take place within it framing the main body of the work. The Usipi, invigorated by the libertas of Britain, may set off on an Ocean adventure. But the whole venture has the ring of implausibility; indeed it is described as being 'like a story' (ut miraculum). It was wonderful in more than one sense. Miraculous if it could be accomplished, but possibly also too astonishing to be true. Similarly, the story concerning Agricola's ship and his lost opportunity for gaining the province of Syria was suspected as a Domitianic fiction. 'It was unclear whether it was true, or whether it was fictitious and made up in accordance with the emperor's character' ('sive verum istud, sive ex ingenio principis fictum ac compositum est', 40.z).52 The events of the Ocean world, especially those which take place on the sea itself, partake of a certain fictive quality. As before, it is worth noting the startling echoes of Germanicus' abortive exploits in the region. The stories brought back by his ship-wrecked men of Britain were miracula (Ann. 2.24). The search for conquest in the form of knowledge cannot be said to be over for Rome, casting some doubt on its grasp on the island and leaving its ambiguous status intact. It is this lack of resolution which gives the Agricola an on-going interest. Britain is left in a state of semi-detachment. Agricola's conquest had apparently destroyed its insularity, but the fictive quality of the final episode alludes back to the Oceanic world which had characterized Britain at the start, and raises questions over the completeness of the Roman mission to embrace the island within its grasp. This might then be jl A similar disclaimer relating to discussion of the j2 It is striking, given other similarities between Ocean is given by Strabo 2.3.3, referring almost Agricola and Germanicus, that, when Germanicus is without question to Posidonius. If, as Ogilvie and recalled by Tiberius, the reason given (namely to give Richmond suggest, Tacitus knew Posidonius' works, Drusus a share in the glory) is considered by Ger- then it is quite possible that he was entering here into manicus fictitious Gngi: Ann. 2.26). a well-worn topos of avoiding overlap with Posidon- ius' monumental work On Ocean. formulated as a Roman failure -Britain has remained an island after all, but only by the skin of its teeth. However. the end of the British narrative does not mark the end of the work. lust as Britain wins a partial reprieve, so too does Agricola's reputation receive a final rehabilitation. I have argued above that there might not be much to commemorate in the res gestae of someone who destroyed the world which encapsulated the values he held dear, or perhaps in the career of a commander whose plans were continually thwarted. However, the final chapters of the work, together with the opening chapters, form a frame around the British narrative, and are similarly concerned with matters of commemoration and fame. I suggested above that Britain, by its very remoteness, could be identified as a place of invigoration, where people such as the Usipi are able to perform in a spectacular and memorable way, and where Calgacus fulfils some of the qualities which Agricola himself might have accomplished, had he not been a general under Domitianic Rome. By contrast, Rome is antithetical to both the performance and the celebration of noteworthy deeds, as is very clear from the opening chapters. Indeed Agricola goes so far as to try to sink into obscurity on his return-to ~ome. An article on the shifting geographical conceptions of Britain in the Agricola is not the place for a detailed exploration of the non-British parts of the work. It is, however, worth noting that the closing chapters pay extensive tribute to Agricola and to his res gestae, and above all secure his lasting fame through Tacitus' commemoration. This must have some bearing on our reading of the central core of the work. namelv his exploits in Britain itself: I suggested th; Britain, as a part of the peculia; and remote Oceanic island world, and therefore free from the corruption of Rome, was the ideal location for Agricola, even though his campaigns threatened to destroy that insular id en tit^.'^ But the continental attachment was never auite com~leted. and Britain retained some elements of an almost fictive Oceanic image, as revealed by the episode which concludes Agricola's dealings with that part of the world. Agricola's rather muted fame is more appropriate to the real world of Roman politics under D~mitian.'~ In answer to the puzzle of how to be a good man under a bad emperor, the opportunities offered by somewhere as distant as Britain are perhaps not sustainable for long, at least not for a Roman. In that context, Agricola's moderation is greater cause for celebration than the abandon of the Usipi or the exuberance of Calgacus. Agricola embodies a new tvDe of ~irtus.~' The threats that are made to Britain's island identitv in the course of the tkft allow for an ex~loration of where the true limits of oma an it as and libertas lie. But the sense of difference and remoteness which my opening quotes would attribute to insular Britain means that that location cannot match or be matched by Rome. There is, after all, 'a sense of place about islands -even when they're semi-detached'. St Hilda's College, Oxford j3 R. >'I.Ogilvie, 'An interim report on Tacitus' jj On the new political virtue of quietism and mod- "Agricola"', ANRW I1 33.3, I I 17-18, argues that eration, see \V.Liebeschuey, 'The theme of liberty Tacitus deliberately belittles the achievements of in the Agricola of Tacitus , Classical Quarterly 16 Agricola's predecessors in Britain so as to increase (1966), 126-39. It has, however, been argued by T. J. Agricola's stature. Luce, 'Tacitus on "History's highest function": prae j4 See C. J. Classen, 'Tacitus -historian between czpuum munus annalium (Ann. 3.65)', ANRW I1 33.4, Republic and Principate', ~Wnemosyne41 (1988), 2904-27, that Tacitus is not, by praising characters in 93-1 16, for the realism of Tacitus' vision; also for the his works, necessarily setting them up as models for argument that even the definition of what constitutes emulation in the manner of Livy. Tacitus is far more moderatio has changed since the time of Cicero. interested in commemoration than in paradigms.

  • Recommend Us