CHAPTER 22
FRANK RUSSELL
ON a wall in the Ashmolean Museum in Oxford hangs a small funerary plaque, consigning to the underworld the shade of Valerius Paternus, speculator of the army of Britain.1 We know little of who he was, other than that he had friends who thought well of him. Likewise, we know little of what he—as a speculator—did, other than scraps and allusions scattered over centuries, which are often at variance with each other. Was he a spy, a courier, an executioner, or a staff officer? Or was he all of these things?
Such questions may seem trivial in the particular, but are certainly relevant in a general application to the wider world of classical intelligence. For to some extent the identities and practices of the people who gathered information influenced the reception of the messages they sent. A comprehensive understanding of the role of intelligence in the ancient world requires context beyond the scope of a chapter—in effect, the history of the interactions of peoples, their polities and ideologies, the structure and degree of professionalism of their military, and their social and economic systems. Such an exercise is worthy of a Thucydides or Livy—herein the task is more like that of Arrian in his Tactics against the Alans: to address particular details and practical concerns. The focus of this chapter is tactical intelligence—the daily and transient needs of the general on campaign or defending a frontier.
Categorization is difficult, for the Latin and Greek vocabulary is inconsistent, not merely between inscriptions and historians, but even within a single author. Therefore the structure of this chapter follows a division by posture: offensive and mobile on the one hand, and defensive or localized on the other. That is not to say that there is no overlap—exploratores based at forts on the Roman frontier resemble in many ways those preceding an expeditionary force—but that the distinction is meaningful and useful.
OFFENSIVE: MOBILE WARFARE AND RECONNAISSANCE
Among both Greeks and Romans, many—but not all—commanders were wont to send armed men ahead of their armies to discover the lay of the land, or the location and disposition of the enemy. All such men were called kataskopoi by Greek authors, a term more or less translatable as “scouts” (also, e.g., skopoi, proskopoi, proodoi). In some instances these would make up strong forces numbering in the hundreds or even thousands, in others smaller units of thirty or so, at still other times small teams, typically of three men. The size of the contingent would be primarily dependent on the relative emphasis on combat: the larger the force, the greater the capacity to engage the enemy, and the more overt (and obvious) the collection; the smaller, the more the unit sought to avoid contact, and the more covert the collection. At the extreme of the range, single individuals—or groups of two or three—might even don disguises or assume cover stories and infiltrate themselves among their foes.
Large-force Reconnaissance—Prodromoi, Procursatores, Excursatores
Armies on the march were often, but by no means always, preceded by a substantial vanguard of light cavalry or, less commonly, light-armed troops. Although these performed reconnaissance functions—by discovering routes, uncovering ambushes and the like—they were not exclusively (perhaps not even primarily) used in that role. Rather they were simply combat units assigned a role that mixed reconnoitering, raiding, and skirmishing. Their nomenclature and practice attest to their reliance on mobility and on visual contact with the enemy. They are typically commanded by officers of some status, who communicate with their general through messengers, or upon their return. The range at which they operated is rather difficult to assess—it is common for them to be in sight of the main force, yet they may be found as far as seven or eight miles, beyond which distance it becomes too difficult to support or communicate with the main force (Russell 1999: 21–2; Austin and Rankov 1995: 44–5; Ezov 1996: 73–4).
Greece
While there are a few examples of the use of a substantial skirmishing and scouting force preceding the main strength of an army or navy in the fifth century, there is a marked increase in the early fourth (see Russell 1999: 11–15; cf. Pritchett 1: 127–33, Spence 1993: 141–51). The Boeotians, who could field a considerable cavalry force (1,100 horse), may have used part of it as a screening and reconnaissance force in the opening moves of the Battle of Delium in 424 (Worley 1994: 94–6). The joint Athenian-Corinthian cavalry force sent by the strategos Iphicrates to locate the Theban army near Mount Oneum (370 B.C.) would have been quite large if both allies had indeed set out with all their strength (Xen. Hell. 6.5.49, 51–2). Athenian cavalry were widely admired for their delaying action against the Thebans just prior to the battle of Mantinea (362 B.C.), which they fought in the absence of their slower-moving compatriots on foot. So too did the cavalry contingent raised by King Agesilaus bask in glory for sending off the Thessalian horsemen; more repute was to be found in combat than reconnaissance (see Spence 1993: 143–6, 151–2; Bugh 1988: 146–50).
The campaigns in Asia Minor, culminating in Alexander’s march east, saw the clash of large cavalry screens operating in advance of slower-moving armies. On his march south along the Aegean coast, an ile of his Companions accompanied his four ilaiofprodromoi, but the presence of the elite Companions was exceptional. On other reconnaissance operations—the prelude to the Granicus and Gaugamela (Arr. Anab. 1.13.1, 3.7.7)—the Companions joined the scouts only after the Persians were discovered, to initiate an attack. The prodromoi also teamed up with light-armed troops and mercenary cavalry for operations, which included pitched battle alongside the elite forces, since they were armed with lances (hence their alternate name sarissophoroi). They proved exceptionally useful and flexible, yet remained in a lower social status than the rest of Alexander’s cavalry, there being some doubt as to whether they were even Macedonian or Thracian (Milns 1981: 351; Brunt 1963: 27–8; Rzepka 2008: 51–2 counts them among the Macedonians). Light cavalry continued to be employed for reconnaissance by Alexander’s successors, but never again with such a strong sense of unit identity. The only other analogous force—used in both reconnaissance and line of battle—was the Sciritae. These tough and wary people living on the northern frontier of Laconia were neither citizens nor slaves, but “dwellers about.” They fought either as hoplites (e.g., at Mantinea, 418 B.C.) or light infantry. To them was assigned guard duty by night and scouting by day. Xenophon (Lac. 13.6) noted that the Lacedaemonians were habitually preceded by the Sciritae and outriders (proereunomenoi hippeis) on the march, which could be a substantial force, given a number of up to six hundred for the Sciritae alone.
Rome
Modern scholars do not generally characterize the Romans as attentive to reconnaissance in the early and mid-Republic, and it must be admitted that their armies on the march were vulnerable to ambush. At Trebia we find Roman cavalry and light-armed troops engaged in the initial clash, but their scouting role is implicit at best; on the march toward Cannae, the heavy infantry received the first shock of the Carthaginian skirmishers, while the Roman cavalry and light-armed troops joined later, which is not the order one might expect if an advanced force was being employed. Even when we find vanguards mentioned in our sources, they seem to find themselves embroiled in combat at the expense of their reconnaissance role. We read, for instance, that during the Second Punic War two consuls took about two hundred cavalry and thirty light infantry (velites) to reconnoiter, only to perish in a disastrous ambush. While Polybius (10.32.1–12; see also Livy 27.26.11) commended their courage, he condemned their folly. The light infantry and cavalry leading the armies of Flamininus and Philip V were no more circumspect in their initial contacts near Pherae (197 B.C.): forces of about six hundred on either side clashed in prolonged skirmishing (Polyb. 18.19.9–12). After two days, during which time the armies had lost contact with each other, and were separated by a ridge and heights, large reconnaissance forces (one thousand light-armed and three hundred horse on the Roman side) blundered into each other in the fog, and the skirmish escalated into an unexpected full-scale battle that was decided as much by chance and terrain as generalship (Polyb. 18.21–2).
In the example of Flamininus above, we see already in the early second century the use of allied cavalry: two troops (oulamoi) of Aetolian cavalry under their own commander led the initial contest at Pherae, and again took part in the early stages of the battle on the heights of Cynocephelae. These were included, says Polybius, because they were familiar with the countryside. Later in the Republic, large numbers of allied cavalry were wielded by Caesar when monitoring the Helvetii—apparently the full force of four thousand, raised from Transalpine Gaul, the Aedui, and their allies. The mass of horsemen seems to have operated more or less as a single body, since Caesar remarked that a much smaller contingent of about five hundred Helvetian cavalry took advantage of favorable ground to rout his men (Caes. B Gall. 1.15.1–2). The setback was by no means decisive, and Caesar continued to use his cavalry to shadow the Helvetii, but with instructions to refrain from combat. While Caesar made better use of cavalry than most (contrast Crassus at Carrhae), Cagniart (1992: 74–5) has acutely observed that he did not integrate them into the Roman military system. Not only did they fight in their own way under their own ethnic leaders, but they did not even maintain their unit identities, since they normally dispersed during winter, and were recruited again in the spring.
Allied cavalry continued to be enrolled during the Empire, in a more permanent and institutionalized manner, when large auxiliary units of five hundred or one thousand horse were available for use. As the Roman Imperial army developed a stronger cavalry force, so did its capacity for using strong mobile vanguards and flanking units. In Arrian’s Against the Alans (1.1–2), for example, the column is led first by kataskopoi, then a number of cavalry alae and cohortes equitatae, totaling in excess of 3,500 men, while another cohort of 500 patrols the flanks. Julian fielded 1,500 cavalry ahead and to his flanks (Amm. Marc. 24.1.2) during his invasion of Persia in A.D. 363. In this instance we can catch a glimpse of them in action in smaller formations soon after—the Persians surprised three turmae (about one hundred men in total), killed a few together with one of their commanders, and captured a standard.
Small Force Reconnaissance—Kataskopoi, Exploratores, Speculatores
If a large, mobile force precedes an army on the march, it may well be that they come to see, but they also come to be seen. A long line of cavalry stretched out over a plain, raising clouds of dust as it rumbles forward, is hardly inconspicuous, and if there are many examples of cavalry screens discovering hostile forces, there are also instances when their presence in turn broadcasts the news that an army is near (e.g., Arr. Anab. 3.7.6–8.2). Smaller units may not be able to hold their own in battle nearly so well, but they are far less likely to betray a move or commitment. These typically operated in platoon-sized groups by day, if exploring open terrain on horseback, especially ahead of a vanguard, or squad-sized teams (from three to ten) in more stealthy roles, by night, on foot, or hidden in observation posts on the heights.
Greece
From the time of the Iliad, reconnaissance by small groups of men, especially by night, has been seen as an especially risky business (Hom. Il. 10). Consequently military writers emphasize that scouts must be intelligent and experienced (Russell 1999: 20–1, 24, citing Xen. Cyr. 5.3.56ff., Aen. Tact. 6.3, Veg. Mil. 3.6). Yet oddly enough, outside of epic, the scouts we meet resemble Dolon more than Diomedes. A case in point is the team, which consisted of two of his servants and a Sicyonian exile, sent by Aratus, laterstrategos of the Achaean League, to investigate a vulnerable point in the walls of Sicyon. These men were quite thorough in providing operational intelligence—measurements of the wall and city defenses, some details about noisy dogs—that enabled Aratus to infiltrate the city (Plut. Arat. 5.3–5). Now it is true that the exile had been over the wall in question before, and hence could count as experienced in a specific way, but these three were hardly professional. Xenophon (An. 4.4.15–16) mentions that the Ten Thousand repeatedly appointed Democrates son of Temnus to lead reconnaissance patrols, since he “enjoyed the reputation of having made accurate reports in many previous cases of the same sort, describing what were facts as facts and what were fictions as fictions.” Democrates does not appear elsewhere in the Anabasis, held no command among light infantry or cavalry, and there is no reason to suppose that he was otherwise of note. He simply happened to be reliable, hence he became experienced.
If we can extrapolate from this example, there was probably a de facto specialization among commanders and units that included reconnaissance in their repertoire, and hence we would do well to look at the cavalry of Athens, for whom we have some epigraphic and narrative testimony. First, it should be noted that cavalry are not all created equal, and the elite are not typically found in small-scale reconnaissance roles, even in democratic Athens. That task was delegated to light cavalry: the horse archers of the fifth century, and the prodromoi and Tarentines of the fourth (Bugh 1988: 39–40, 221–4). Spence (1993: 134, 149–50) holds that the main role of the prodromoi was reconnaissance, and combat was secondary. He may well be correct, especially if they numbered fifty (five from each tribe), rather than two hundred, as did the horse archers before them. Like the mounted archers they did not belong to the cavalry social class. They were not wealthy enough to supply their own horses, and were maintained by the phylarch. Given this support, they may have provided small-team reconnaissance for their patron phylarch, and hence would have set out singly, or in groups of two to five. Spence also suggests that they may have acted as couriers, or aides (hyperetai), a point which could make an interesting connection to Xenophon’s recommendation (Eq. mag. 4.4) that the hyperetai ride in advance of each tribal regiment (phyle) to find routes through rough territory, be it friendly or hostile. They are further associated with light infantry (hamippoi), who at times accompanied them into battle, and they are closely connected to the cavalry clerks (grammateis) and commanders, emphasizing their intelligence function (Bugh 1988: 222ff., following Elkman 1933; Bugh 1998: 86ff.; Spence 1993: 150–1).
In addition to their roles in exploring terrain and locating enemy forces, reconnaissance teams were fielded to prevent surprise. As mentioned earlier, large advance guards tend to attract attention, and hence provoke ambushes. While it is all very well and good to have your enemy tip his hand and reveal his forces—that is what a vanguard is for—it is better still to have advance warning and win or avoid the confrontation. Therefore we find Greek and Roman strategists posting scouts in front of scouts (e.g., Xen. Eq. mag. 4.5; Veg. Mil. 3.6); a colorful example is the scheme of Xenophon to send men disguised as bands of brigands in both number and garb, who would either capture anyone who might report their presence or, if the informant escaped, would provoke only such a counter as would be thought appropriate for a band of thieves (Xen. Cyr. 2.4.23). Arrian (Against the Alans 1.1ff., with DeVoto 1993: 121) calls the first group in his order of march kataskopoi, which are followed by light cavalry units. Arrian does not indicate the number of kataskopoi, nor make a point of separating them from the rest of the column until they reach the battlefield. At that point, they deploy onto the heights to observe the enemy, rather than remain on the plain with the rest of the cavalry. On the one hand, thekataskopoi have their own commander and distinct mission—one that the light cavalry could have fulfilled—and so it is possible that they are a unit of exploratores. On the other, Arrian does not give them a specific unit name (in contrast to other formations, like the mounted archers from Petra, the Cohors III Ulpiana), which leads one to wonder how much unit identity is warranted. Pseudo-Hyginus, in a tactical work of the third century A.D., includes two hundred exploratores attached to the considerable force (three legions plus a large cavalry presence) that he describes in his Camp Fortifications (30); like Arrian, he gives a round number of individuals rather than units. The lack of specific references in these Imperial texts echoes a problem that stretches back to the Republic: to what extent was there an organizational structure superimposed on reconnaissance?
Still more fundamentally we find ourselves once again with a problem of definition. Small-scale reconnaissance in Latin texts is typically the province of men called exploratores or speculatores. What exactly these terms mean, and the distinctions between them, is not always certain. The general consensus is that (a) the terms do not have technical significance before the first century B.C. (at earliest), and (b) the distinction and peculiarity of the titles become more definite during the Imperial period (cf. Ezov 1996: 72, 81–3, 85–6, 90–3).
The term explorator seems to have evolved from a general label for any soldier engaged in reconnaissance to a specific military classification for a type of soldier. It is first attested in the late Republic (although later authors use it in the context of events as early as the third century B.C., e.g., Frontin. Str. 1.1.6). Caesar used the word in his own writing to denote scouts, although it is probable that he meant no more than the Greek “kataskopos”—that is, the term defines the action rather than the unit. This can be inferred from any number of examples. Caesar ordered the allied Ubii to send exploratores into the country of the Suebi (B Gall. 6.10) and it is apparent that this is a generic term, since the organization of the Ubii, such as it was, was entirely native. Also, when Caesar sent C. Volusenus into Britain, he chose him on the grounds that he was suitable (idoneus) for the task of a protracted reconnaissance. Yet Volusenus was a military tribune with no connection to the exploratores or speculatores; if there was a standing body of such men, it would have made sense to choose among them (Caesar B Gall. 4.20–21; Austin and Rankov 1995: 100–1; cf. Gichon 1989: 158–9).
Perhaps as early as the first century A.D., and certainly by the second, the exploratores were more formally organized. Gichon (1989: 159) leans toward an earlier date (the Principate) while Austin and Rankov (1995: 189–95) are more cautious. Gichon goes on to note (1989: 159, citing Ps.-Hyginus camp fortif. 24, 30) that by the time of Hadrian “they were quartered near the 1st cohort, the equites legionis and the singulares. They are thus marked as a special unit for immediate disposal by the legatus whenever need arose, similarly to the position of our day combat intelligence and reconnaissance units in battalions and higher formations.” Austin and Rankov (1995: 190–1) cogently argue that there was a development from detachments composed of men seconded from their original (usually auxiliary) formations, to permanent standing units that were either derived from temporary units or expressly raised as exploratores. They continued to serve throughout the Empire (being particularly extensive and well attested in Upper Germany), in small units attached to infantry or mixed cohorts, or in forts along the limes (Speidel 1983; Austin and Rankov 1995; Southern 2007; Gichon 1989: 159–60, 162–5).
This evolution may have given birth not just to a new species but a new class of unit, and the formations of exploratores posted along the limes will be treated below. For the moment, it is worth noting that there is a distinct divide between the epigraphic and narrative testimony for exploratores. On the one hand, we find a fair number of numeri and cohorts of exploratores of up to a thousand strong attested in inscriptions. Yet Arrian and Ammianus Marcellinus, in the second and fourth centuries A.D. respectively, make no mention of any unit of exploratores in conjunction with a mobile force but rather refer to them in a manner strikingly like Caesar did in the first century B.C.—that is, vague references in the plural (see also Austin 1979: 118–19). Line units are given name and number, reconnaissance teams are anonymous.
Nonetheless, information about their operations is not entirely absent. When numbers are given, they range from a few to a few dozen (Austin and Rankov 1995: 192–4, Dixon and Southern 1992: 31–2) and may comprise both cavalry and foot. They operated well beyond the typical screen of outriders (Austin and Rankov 1995: 44–5; Ezov 1996: 74, giving examples of up to about twenty-four miles), and communicated with their generals by couriers, or upon their return.
The general conduct of reconnaissance changed very little over the millennium that stretched from the age of the classical polis to the late Roman Empire, and varied according to the resources of the power and the terrain rather than differences in polity or technology. True, there was an increase in the use of vanguards among the Greeks after the fourth century B.C., and among the Romans in the first. Cavalry was widely used in this role, but there was a transition from elite citizens in the age of the polis and early Republic to a heavy reliance on allies in the late Republic and Empire. Smaller-scale reconnaissance never became a specialty of the Greeks, but eventually became the task of specific types of soldiers, even of specific units, in the Roman period. Leadership of reconnaissance tasks, however, was not always entrusted to a designated unit commander, who was frequently superseded by a more trustworthy (or at least more influential) officer—at times even the general, king, or consul himself.
DEFENSE: SURVEILLANCE OF A FRONTIER
The fixing of a fair number of units of exploratores at specific locations along the frontiers during the Imperial period leads to the general question of the role of reconnaissance in border security. There are some differences between mobile reconnaissance, as described above, and essentially static surveillance: the actors may be the same, but the roles, names, organization, and method are not. Testimony for an organized watch on a frontier extends all the way back to the Mycenaean period, as Linear B archives from Pylos attest (Russell 1999: 24–5, citing Ventris and Chadwick 1956: nos. 56–60).
Towers blossomed on the heights of Greece in the classical era, flourished through the Hellenistic period, and faded in the Roman. They were particularly abundant in the northern Peloponnese, the southern mainland, and in the Aegean islands. Whether or not they were built primarily as observation posts, most could play the role, since they tended to be sited on hilltops with an extensive view. Often the view included other towers with which signals could be exchanged (Lawrence 1979: 187–9), and several scholars have described chains of towers and forts along the Attic frontier. These provided not only local security, but also a sort of early warning system designed to relay signals along one or more lines of sight back to the city (e.g., Munn 1993: 94–5). Camp (1991) has identified remnants of what might have been a parallel system on the Boeotian side, and inscriptions from other parts of Greece suggest that Athens and Thebes were by no means singular in this respect (e.g., Bean and Duyuran 1947 on Smyrna; Ma 2000 on Miletus).
The maintenance of the watch on the frontier was of particular concern to the state, which directed and funded it, and enrolled men to staff it. Commanders were appointed (typically by election in the poleis) to ensure that garrisons of the forts were maintained. These men tended to be of stature, and in some of the smaller cities of Asia Minor the office was eponymous (i.e., the holder’s name identified the year, cf. Sherk 1991, s.v. Clazomenae and Bithynian Cius). A third-century B.C. inscription from Teos describes the duties of one of these commanders (here called a phrourarkhos, SEG 26.1306). After his nomination and subsequent election, the phrourarch swore an oath, witnessed by the strategoi and magistrates, to discharge his duties faithfully and to remain in the vicinity of Teos, under pain of death.
The strictures and punishments were particularly exacting due to the fear of a revolt and the potential havoc that could be wrought by a force commanding the heights. So too were the qualifications: only citizens who have dwelt in Teos for at least ten years need apply. While in the khora, the phrourarch received four Alexandrian drachmae per day from the public treasury, while each of the citizens serving under him earned a single drachma. He was expected to maintain at least twenty men at each fort (the number of forts is not specified), who were armed as light infantry—with shield (aspis), helmet, spear, and knife. These men were obliged to obey him or suffer dismissal without pay. The phrourarch was also to maintain the three dogs that the state provided for each post, no doubt to alert the men to intruders by night. Dogs are also mentioned in other decrees, and there are colorful and even comical stories of infiltrators trying to elude or thwart the canine companions of night patrols along the city walls (Russell 1999: 29–31).
The arrangements for Teos reflect the general measures among the Greek poleis. The phrourarch’s term of office typically ranged from four months to a year—some states did not regulate reappointment; others mandated an extended period between terms of office (e.g., five years, SEG 41.1003). Command of the Attic frontier was divided between two strategoi after around 261/260 B.C.: one holding the western, the other the eastern half of the defensive line. These posts could be held by the same person on more than occasion. The watchtowers and forts were occupied at least part of the year by garrisons, and served as bases for patrols (peripoloi) that were drawn from them, or undertaken by cavalry. Kent (1941: 347) has argued that, in the fifth and early fourth centuries, the watch on the Attic frontier waxed and waned according to international tension, and passages in Aeneas Tacticus (22.26) and Demosthenes (De cor. 37–38) confirm that more attention was paid to night watches, at least in times of crisis, implying that vigilance relaxed during the comfortable interludes of peace. On the other hand, bandits were never altogether absent from the frontiers, and the task of policing man and beast was a year-round commitment from the mid-fourth century to the early first.
In the fifth century the men who kept watch were usually citizens; as time progressed they were joined or replaced by mercenaries. For the most part, the mercenaries were hired to relieve citizens of the burden, but in Athens Macedonian mercenaries and Athenian citizens shared the watch on the frontier during the period of Macedonian domination (261–229 B.C.). While some forts, like Eleusis, could accommodate perhaps a hundred men, most were more comparable to the Tean posts of twenty or so. In practice, garrison and patrol duty often fell to the young men. In Athens, for example, the ephebes were assigned this task, which they undertook for a period of two years, until 305, when the term was reduced to a single year (Bugh 1988: 185 n. 4).
ROMAN LIMES
The expansive and expanding limits of Roman influence became defined more formally and elaborately in the Imperial period. While significant regional variation endured, the days of ad hoc governance faded with the ideals of the Republic. Provincial administration was increasingly entrusted to a structure of oversight built on foundations in Rome and regional capitals. In the latter were located the officia of the emperor’s legates, complete with a support staff of professionals—cornicularii (executive officers) with their assistants, commentarienses (judicial recorders), speculatores, frumentarii (logistical and police officers), quaestionarii (interrogators and torturers), interpreters, and the like—even a haruspex (diviner). These would in turn be supported by a secretarial staff, which recorded and processed the information necessary to maintain, supply, and field a complex professional army. The legions commanded by the legates were the concentrated striking force of the Empire, deliberately removed from the day-to-day watch on the frontier. That task was delegated to detached cohorts and numeri, more often than not populated by the many ethnic groups that served the idea and the paymaster that was Rome. For the purposes of exploring the question of tactical intelligence, it is possible to divide the units along the limes into two categories: fighting formations that fielded reconnaissance patrols and manned observation posts, and specialized reconnaissance units.
The larger formations, such as the cohorts of five hundred or a thousand, were often dispersed during times of peace. We are fortunate to have a number of their duty rosters (pridiana), which mention detachments on scouting assignments: in the case of duty rosters of the Cohors XX Palmyrenorum (dated to A.D. 219 and 222), ten horsemen and five infantry are recorded on a scouting mission (Dixon and Southern 1992: 135; Gichon 1989: 162 n. 33; Webster 1985: 150 n. 2); the pridiana of the Cohors I Hispanorum Veterana quingenaria equitata, dated to around A.D. 105, lists a cavalry patrol beyond the Danube (BM Pap. 2851 ii.32, in Fink 1958: 104). Starting from the late second century, more formally constituted units of perhaps twenty to thirty exploratores might be stationed with a much larger force, as was the case at Welzheim-Ost and Walldürn on the German frontier (Southern 1989: 103–4, 111–12), and at Bremenium and Habitancum, located beyond Hadrian’s Wall (respectively modern High Rochester and Risingham, two locales separated by less than fifteen kilometers in Northumberland). The inscriptions testifying to the presence of the units in Britain are dated to the early third century (A.D. 238–241 and 238–244, respectively) and show them in association with mixed infantry and cavalry cohorts of one thousand strong.2 Given that a cohors equitata was suited by nature to conducting reconnaissance by itself, as we have seen in the duty rosters, the presence of the exploratores suggests a complementary role (perhaps longer-range excursions) or recognition of a need for specialized troops, or both. There is some debate whether the exploratores and the numeri were entirely separate—they are mentioned separately in inscriptions—or whether the exploratores were integrated into the largernumerus.
While some exploratores were on temporary service detached from the auxiliary formations (e.g., from a cavalry ala, Southern 1989: 110 citing CIL 8.21516, 13.11735), specific units designated as exploratores appear in the late second century. In some cases an existing unit was renamed, for instance the Brittones Triputienses at Miltenburg became the Exploratores Triputienses (Southern 1989: 108), and the Cohors IX Batavorum equitata milliaria in Raetia added expl. to its title (Southern 1989: 111). One can only presume that the redesignation was occasioned by demonstrated aptitude or experience in reconnaissance. Other units, such as the Ala Exploratorum Pomariensium in Mauretania Caesariensis and the Numerus Exploratorum Divitiensium at Niederbieber may have been raised as specialized units. These units were mobile, and often mounted (in the oft-cited case of the exploratores attached to the Cohors XX Palmyrenorum, on camels). While they might have occasionally gone out in full strength, it is probable that they more typically operated in small teams, or even singly.
Very large units, like the exploratores Divitienses, were dispersed over more than one location (Southern 1989: 104). Smaller units, like the Triputienses, Halicanenses, and Triboci et Boi, may have recruited locally, and have been quite familiar with the geography and customs of their allotted regions (Southern 1989: 114). It is generally agreed that exploratores operated well beyond the limes, at distances up to about 35 km (Austin and Rankov 1995: 44–6; Gichon 1989: 157, 164–6). As they became established for substantial periods at fixed spots on the periphery they came to resemble a more professional version of the Greek peripoloi, providing a flexible and versatile complement to the static watchtowers along the limes. They were one of the faces of Rome on the frontier, and established long-term relationships with the neighboring peoples of the barbaricum, even as they monitored the local mood and movements. Their information would flow back through messengers or reports in the relative calm that typified daily life in most areas, but when the storms of crisis broke the outposts could flash signals from point to point along the frontier (see Woolliscroft 2001).
If exploratores are occasionally difficult to define, speculatores are positively elusive. Their ambiguity is heightened by the fact that, from the first century A.D., the title is given to a number of roles (messengers, members of a governor’s praetorian guard, and even executioners) connected only tenuously by a general association with information (Austin and Rankov 1995: 54, 150; Rankov 2006: 130). Even within Caesar’s texts there is some ambiguity—the noun speculator, for example, may be found with the verbexplorare (e.g., B Gall. 5.49.8)—but the general tenor is that they are agents who operate more covertly than exploratores. In some episodes they appear seeking the same sort of information as reconnaissance teams, but do so under cover of darkness (B Gall. 2.11.2–3). In discussing Caesar’s account of a nocturnal departure of the Belgae and Bellovaci, scholars (Austin and Rankov 1995: 56; Ezov 1996: 80; Gichon 1989: 157–8) find a clear differentiation between the speculatores, who send reports of the move from the time it began, in the second watch of the night, and the exploratores, who confirm the reports after daybreak. On another occasion, Pompeian speculatores in Spain were captured by Caesar during the night (B. Hisp. 20.5); a while later (B. Hisp. 28.1), we read that Caesar’s speculatores were more successful, and sent reports on Pompey’s army activities from the third watch of the night. A similar situation is implicit in Ammianus (31.11.2) who, writing some centuries later, mentions that speculatoresapproached close to the campfires of the Goths, and heard the noises of man and beast. In other contexts, we find allusions to speculatores in disguise, or pretending to be deserters or diplomats (e.g., Caes. B Afr. 35.2–5; Austin and Rankov 1995 ad loc.). Yet we can also find them approaching the outworks of an enemy camp either as or amid mounted scouts (B. Hisp. 13).
Cover facilitates proximity—speculatores tried to get near or among the enemy. They seemed to listen as much as they observed, a curious habit given that their name is linked to vision (contrast the Greek otakoustes, or “eavesdropper”). Thus while the types of intelligence they sought were much the same as exploratores, their method differed. So too did their fate if captured. Unlike cavalry skirmishers or recon teams, whose lot if captured was no different from any other soldier, speculatores were killed—slaves were crucified, citizens beheaded (Span. War 20.5)—even by Caesar, who valued a reputation for clemency during the Civil Wars. The varied executions indicate the varied backgrounds of these sorts of agents, who included Roman citizens, allies, local inhabitants (e.g.,Span. War 38), and slaves. In turn, the varied backgrounds and the ad hoc nature of some of the stories (e.g., Metellus Scipio’s recruitment of Gaetulian allies to pose as fake deserters, B Afr. 35) diminish the strength of arguments for a defined corps. Austin and Rankov do, however, identify a coin of Mark Antony’s that records a cohors speculatorum, suggesting that some structure, even if temporary, was emerging (1995: 189, citing Grueber 1910: 527, nos. 185–6). The question remains what sort ofspeculatorescomprised this cohort. The bulk of the epigraphic evidence locates speculatores either on the staffs of provincial governors, where they served as aides, messengers, and apparently executioners (Austin and Rankov 1995: 150–1; Clauss 1973: 59–77), or among the Praetorian Guard (Clauss 1973: 46–58). It is quite possible that in the former case they had an intelligence role, since the staff was undoubtedly composed to facilitate the governor’s communications and decisions (see Austin and Rankov 1995: 150–4; among others, the staff included ten speculatores for each legion stationed in the province). Speculatores also seemed to have a peculiar identity—there are a number of clubs attested in provincial capitals (scholae speculatorum, see Clauss 1973: 158–9, nn. 100–01, 105, 108, 131) which suggest an exclusive and collegial relationship, perhaps even a professional one.3
While the bulk of the evidence leads to the conclusion that the speculatores attached to the legions left the field for the office sometime in the first century A.D., Gichon (1989: 167, citing Baradez 235–42) draws attention to forts along the Numidian frontier that were bases for speculatores, who may also have operated signal towers on heights just above their location. We may conclude, therefore, that not all were posted at headquarters, and perhaps that even those who attended the governor could be used for covert collection. Conversely, we may wonder if there is confusion between those who held the title of speculator (either on the governor’s staff or in the Praetorian Guard) with those who were called speculatores because of their function as spies; certainly such ambiguities have parallels in the Roman world. One can find a pertinent analogy in the frumentarii, whose name recalls their original function as logistics officers, but who could equally be classified as staff officers in a governor’s officium or as secret police who monitored domestic and political threats (see Sinnigen 1959, 1961, 1962, 1964; also Clauss 1973: 82–113). Austin and Rankov (1995: 137) are probably right to dismiss the idea that the frumentarii contributed to the collection of military intelligence at the frontier, but their association with the speculatores at headquarters is as curious as their nickname (curiosi). Also, at some point after Diocletian disbanded the frumentarii, their domestic intelligence function was acquired by the civilian agentes in rebus, while the arcani orareani operated beyond the borders to provide advance warning of threats; both were notoriously corrupt (Gichon 1989: 169). While much of late Roman espionage is inferred from, rather than informed by, narrative sources, Lee (1989a and b, 1993) has done a fine job reconstructing the flow of information across the frontier, and assessing the credibility of Procopius’s allegation that Justinian neglected to fund his spies, and so was vulnerable to Persian moves.
GREEK ESPIONAGE
Whereas Roman intelligence, both covert and overt, became relatively organized and directed by the state, the Greek poleis tended to delegate quite a bit, whether officially or not, to magistrates and generals. Like the Roman speculatores of Caesar’s day, Greek spies resembled reconnaissance agents in some ways; indeed, while there are specific names for various types of spies, they all can share the label kataskopos with their more overt brethren (for spies in Greece generally, see Richmond 1998; Russell 1999: 103–39).
Like reconnaissance, espionage aimed at the particular goal of discovering the enemy’s numbers, location, and formations. On the other hand, spies did not focus on topography or routes, but rather on capabilities and plans, and occasionally details such as passwords. Their time in the field was longer, and they ranged farther, than other agents, and they often used a cover story together with an innocuous appearance (such as a merchant), or even a disguise. Curiously, spies are found in small groups more frequently than alone; three is a common number (see Russell 1999: 135 n. 102, 137 n. 106). According to Polyaenus (5.33.6), Pompiscus, an Arcadian strategos, made a habit of choosing three men unknown to each other in order to avoid collusion, and so that he could interview them separately and cross-check their reports. Spies usually reported back at the end of their mission rather than communicate during its course. There are a number of notable exceptions, and the fourth-century writer Aeneas Tacticus (31) abounds with clever devices for secret communication, including a rather ingenious dead-letter drop at a shrine (31.15–16).
Xenophon was rather interested in how spies might be useful to a general on campaign, and in the Education of Cyrus (6.1.31–43, 6.3.11–20), he weaves an intricate and charming tale of an ill-fated lover turned spy, Araspas by name, which serves as a lesson in how to prepare and field an agent plausibly posing as a fake deserter (cf. also Eq. mag. 4.8). The kataskopoi whom Xenophon commends are not assigned to conventional reconnaissance, since he specifically advises the recruitment of merchants, neutrals, and fake deserters, who have no place in that world. It is notable that espionage would be pertinent to the hipparch in Athens; perhaps it was not so in practice, or perhaps not delegated only to the hipparch (Nicias, for example, fielded spies in Syracuse, although these seemed to owe allegiance to him rather than to his office or the Athenian state; Russell 1999: 131–2, 138–9).
Similar coincidences existed at Sparta. A passage in Herodotus (1.67) tells of Lichas, a Spartan spy sent into neighboring Tegea. The details of the tale are more myth than anything else, but Herodotus speaks in the present when he says that five of thehippeis(here, perhaps, more the social class than cavalry per se) are sent out of the state each year on similar tasks. We also find the hippeis and their officers (hippagretai, marshals of the knights) handling the secret arrest and interrogation of Cinadon, a conspirator planning a coup in Sparta (Xen. Hell. 3.3.4ff.). These coincidences suggest, albeit hesitantly, a link between cavalry and espionage, one that may derive from the notion that spies are, at some level, well-disguised scouts.
Greek military intelligence never became professionalized, not even under the Macedonian monarchs, and did not reflect the sophistication of the prototypical organizations fielded by the tyrants of Cyprus and Sicily in the fourth century (see Russell 1999: 107–15). Some have proposed that the Spartan krypteia, and a possible Athenian counterpart, kryptoi (“hidden [men?]”), mentioned in a third-century Attic decree (SEG 24.154), may have served as a covert intelligence organization, but the evidence is far from sufficient to warrant anything but the most tenuous hypothesis at this point (Knoepfler 1993; Balcer 1976: 282–3; Russell 1999: 119–20).
While the focus of this chapter has been agents active in tactical intelligence, it would be remiss not to mention—even in passing—other sources. Among the more important were captives who, when taken in skirmishes before a battle, could provide or corroborate information for immediate exploitation, or, when taken after a battle, could help a commander understand what exactly happened, and perhaps even why it happened (Russell 1999: 42–9; Austin and Rankov 1995: 67–73). Both Greeks and Romans were known to send out troops to waylay stragglers or scouts when they needed more information than observation alone could supply. Deserters from the enemy were almost always welcome, and could even be rewarded with wealth and status (think of Josephus), but have always been and still are handled with caution, for false deserters—who might seek to misinform, spy, or betray—play many parts in our drama (Russell 1999: 49–54; Austin and Rankov 1995: 73–81). Local inhabitants might be questioned, especially about topography and routes, and even brought into service as guides (Russell 1999: 54–60; Austin and Rankov 1995: 81–3). Communication with all of these often required interpreters who, particularly in the case of the Greeks, were usually from a different people than those they served. A prudent commander needed to be able to overcome not only the nuances of language but those of motive: why would a prisoner be only too willing to talk; what would he be likely to conceal; and how could you discover something that you did not necessarily know it was important to seek?
CONCLUSION
The ancient world of intelligence is a world of shades, not so much because of its inherent secrecy, but because it was inhabited by the humbler sort, who were largely anonymous even to those they served. There are exceptions: most notable perhaps is Marcus Vettius Valens, who began as a soldier of the Praetorian Guard, served as trainer (exercitor) of the mounted speculatores, and retired as governor (procurator) of the province of Lusitania. A stunning career, but atypical and due more to his family position than any association with reconnaissance (Clauss 1973: 55; Syme 1939: 244; note that he never actually served as a speculator).
Although there may be merit to Speidel’s (1983) suggestion that at least some units of exploratores were considered elite, Southern’s cautions (2007) regarding evidence are entirely justified. Even if it were true, the general tenor is decidedly otherwise. The Sciritae kept watch while the Spartan peers slept. Young men looked out from lonely hill forts while their elders debated policy in council. The aides of the phylarchs, the mercenaries, the social and economic inferiors of the Athenian and Macedonian cavalry classes, rode through brush, in fear of ambush, so their betters could follow in relative safety. Gauls or Goths, sometimes under Roman commanders, brought news of their kin’s misdeeds to their Roman masters. In such contexts, it was an insult to be called a son of an explorator (Tac. Ann. 11.16), although perhaps the sense of ethnic betrayal diminished as the auxilia came of age.
The peltast, the light cavalryman, and the horse archer were social lightweights. Those who fought in line of battle were the archetypes of valor and civic participation. It is in the press of battle, sang Homer, where the great go to win glory for themselves, or grant it to another in their death. Such valor is conspicuous, whereas a successful reconnaissance patrol is unnoticed. Sentry duty is boring; its trials are quiet miseries, not spectacular sacrifice. An effective watcher requires prudence as much as courage; so too, says Xenophon (Eq. mag. 7.7–8), in reconnaissance, fear is your ally. But, among the rest of mankind, courage in battle was to be admired; wariness and stealth were ambivalent. Perhaps this social phenomenon is the explanation of the otherwise inexplicable neglect of tactical intelligence by many commanders of the classical age. It is not so much that they failed to recognize that information could be useful, but that they assumed that it would work its way up the social chain or, if not, that the lion’s skin had no need of the fox’s. It was the weaker party who needed to know. Why, asked Demosthenes of the Athenians, are you making such a fuss about what Philip is said to be doing or not doing—he should be worried about what you are planning! As Lendon (2005: 4) so eloquently stated, people wage war according to their cultural values. The more successful generals, like Caesar, and innovative thinkers like Xenophon, were willing to push their cultural boundaries and pay attention to intelligence. Still, the fact remains that professionalism and unit identification in intelligence came neither to the poleis nor the kingdoms of Classical or Hellenistic Greece, and came finally to the Romans at least a century after it had pervaded the legions.
In closing, it may be fruitful to return to our beginning beside the memorial to Valerius Paternus. Who was he? He and his ilk were men of middling means. Even as they patrolled the frontier or rode beyond the column, they lived at the margins of society: a youth, a mercenary, an auxiliary from a people in the shadow of some splendid empire. His deeds would not be the stuff of legend, and his name would be far more likely to adorn a humble tombstone than the page of a history book.
BIBLIOGRAPHY
Austin, N. J. E. 1979. Ammianus on warfare: An investigation into Ammianus’ military knowledge. Brussels.
———, and N. B. Rankov. 1995. Exploratio: Military and political intelligence in the Roman world from the second Punic war to the battle of Adrianople. London.
Balcer, J. 1976. “Imperial magistrates in the Athenian empire.” Historia 25: 257–87.
———. 1977. “The Athenian episkopos and the Achaemenid king’s eye.” AJP 98: 252–63.
Baradez, J. 1949. Fossatum Africae. Recherches aériennes sur l’organisation des confines sahariens à l’époque romaine. Paris.
Bean, G. E., and R. Duyuran. 1947. “Ad Tepe again (Sancakli Kalesi).” JHS 67: 128–34.
Bertrand, A. C. 1997. “Stumbling through Gaul: Maps, intelligence, and Caesar’s Bellum Gallicum.” AHB 11:107–22.
Breeze, D. J. 1976. “A note on the use of the titles ‘Optio’ and ‘Magister’ below the centuriate during the Principate.” Britannia 7: 127–33.
Brunt, P. A. 1963. “Alexander’s Macedonian cavalry.” JHS 83: 27–46.
Bugh, G. 1988. The horsemen of Athens. Princeton.
———. 1998. “Cavalry inscriptions from the Athenian agora.” Hesperia 81–90.
Cagniart, P. 1992. “Studies on Caesar’s use of cavalry.” AncW 33: 71–85.
Camp, J. 1991. “Notes on the towers and borders of classical Boiotia.” AJA 95: 193–202.
Clauss, M. 1973. Untersuchungen zur den principales des römischen Heeres von Augustus bis Diokletian. Cornicularii, speculatores, frumentarii. Diss., University of Bochum.
DeVoto, J. G. (ed.). 1993. Flavius Arrianus, tactical handbook. Chicago.
Dixon, K. R., and P. Southern. 1992. The Roman cavalry from the first to the third century A.D. London.
Domaszewski, A. 1890. “Zur Geschichte der römischen Provinzialverwaltung.” RhM 45: 203–11.
Dvornik, F. 1974. Origins of intelligence services. New Brunswick.
Elkman, E. 1933. Zu Xenophons Hipparchikos. Diss., University of Uppsala.
Engels, D. 1980. “Alexander’s intelligence system.” CQ, ns 30: 327–40.
Evans, J. 1987. “Cavalry at the time of the Persian Wars: A speculative essay.” CJ: 97–106.
Ezov, A. 1996. “The ‘missing dimension’ of C. Julius Caesar.” Historia 45: 64–94.
Ferrill, A. 1992. “Roman military intelligence,” in K. Neilson and B. J. C. McKercher (eds.), Go spy the land: Military intelligence in history. Westport, 17–29.
Fink, R. O. 1958. “Hunt’s Pridianum: British museum papyrus 2851.” JRS: 102–16.
Forbes, C. 1930. “Peripoloi at Sicyon.” CP 25: 75–7.
Fournie, D. A. 2004. “Harsh lessons: Roman intelligence in the Hannibalic war.” International Journal of Intelligence and Counterintelligence 17: 502–38.
Gauld, W. 1990. “Vegetius on Roman scout boats.” Antiquity 64: 402–6.
Gerolymatos, A. 1986. Espionage and treason: A study of the proxenia in political and military intelligence gathering in ancient Greece. Amsterdam.
Gichon, M. 1989. “Military intelligence in the Roman army.” Historia: 154–70.
Grueber, H. A. (ed.). 1910. Coins of the Roman republic in the British museum. London.
Kent, J. H. 1941. “A garrison inscription from Rhamnous.” Hesperia 10: 342–50.
Knoepfler, D. 1993. “Les kruptoi du stratège Épicharès à Rhamnonte et le début de la guerre de Chrémonidès.” BCH 117: 327–41.
Lawrence, A. W. 1979. Greek aims in fortification. Oxford.
Lee, A. D. 1989a. “Campaign preparations in late Roman-Persian warfare.” in The eastern frontier of the Roman Empire. BAR: 257–65.
———. 1989b. “Procopius, Justinian, and the kataskopoi.” CQ, ns 39: 569–72.
———. 1993. Information and frontiers: Roman foreign relations in late antiquity. Cambridge.
Lendon, J. E. 2005. Soldiers and ghosts: A history of battle in classical antiquity. New Haven.
Ma, J. 2000. “Fighting poleis of the Hellenistic world,” in H. van Wees (ed.), War and violence in ancient Greece. London, 337–76.
Milns, R. D. 1981. “Asthippoi again.” CQ, ns 32: 347–54.
Munn, M. 1993. The defense of Attica: The dema wall and the Boiotian war of 378–375. Berkeley.
Ober, J. 1985. Fortress Attika: Defense of the Attic land frontier 404–322B.C. Leiden.
Oliver, G. J. 2007. War, food, and politics in early Hellenistic Athens. Oxford.
Papalas, A. 1984. “Dracanum Icaria.” Onomata 9: 55–64.
Pauli, A. 1957. “Prodromoi.” RE 23.1: 102–4.
Rankov, A. 2006. “Les frumentarii et la circulation de l’information entre les empereurs romains et les provinces,” in L. Capdetrey and J. Nelis-Clément (eds.), La circulation de l’information dans les états antiques. Paris, 129–40.
Richardson, M. 1984. Julius Caesar’s use of intelligence. MA Thesis. Defense Intelligence College.
Richmond, J. A. 1998. “Spies in ancient Greece.” G&R 45: 1–18.
Riepl, W. 1913. Das Nachrichtenwesen des Altertums mit besonderer Rücksicht auf die Römer. Leipzig.
Russell, F. S. 1999. Information gathering in ancient Greece. Ann Arbor.
Rzepka, J. 2008. “The units of Alexander’s army and the district divisions of late Argead Macedonia.” GRBS 48: 39–56.
Sheldon, R. M. 2005. Intelligence activities in ancient Rome: Trust the gods, but verify. Portland.
Sherk, R. 1991. “The eponymous officials of Greek cities III.” ZPE 88: 225–60.
Sinnigen, W. 1959. “Two branches of the late Roman secret service.” AJP 80: 238–54.
———. 1961. “The Roman secret service.” CJ 57: 65–72.
———. 1962. “Origins of the frumentarii.” Memoirs of the American Academy in Rome 27: 213–24.
———. 1964. “Chiefs of staff of the Roman secret service.” BZ 57: 78–105.
Southern, P. 1989. “The numeri of the Roman imperial army.” Britannia 20: 81–140.
———. 2007. The Roman army: A social and institutional history. Oxford.
Speidel, M. P. 1970. “The captor of Decebalus, a new inscription from Philippi.” JRS 60: 142–53.
———. 1983. “Exploratores: Mobile elite units of Roman Germany.” Epigraphische Studien 13: 63–78.
Spence, I. 1993. The cavalry of classical Greece: A social and military history with particular reference to Athens. Oxford.
Syme, R. 1939. Review, M. Durry, Les Cohortes Prétoriennes. JRS 29: 242–8.
Tomlin, R. S. O. 1992. “The twentieth legion at Wroxeter and Carlisle in the first century: The epigraphic evidence.” Britannia 23: 141–58.
Ventris, M., and J. Chadwick. 1956. Documents in Mycenaean Greek. Cambridge.
Ward, J. H. 1972. “Vortigern and the end of Roman Britain.” Britannia 3: 277–89.
Watkin, W. T. 1884. “The Roman forces in Britain.” AJ 41: 241–70.
Webster, G. 1985. The Roman imperial army. 3rd ed. Totowa.
Woolliscroft, D. 2001. Roman military signaling. Stroud.
Worley, L. J. 1994. Hippeis: The cavalry of ancient Greece. Boulder.
Zlattner, M. 1997. “Hannibals Geheimdienst im Zweiten Punischen Krieg.” Xenia 39. Constance.