Conclusion
6
Were Troops Really Stationed in Caesarea During Agrippa’s Rule?
Some scholars doubt that Roman soldiers would have been stationed in Caesarea at the time Luke describes in Acts 10 , hence are skeptical of Luke’s reliability on this matter. The prestigious commentary of Ernst Haenchen, followed by some other scholars, suggests that no Roman soldiers would have lived there during Agrippa’s rule (from 41 to 44 CE), and that the event Luke describes must have occurred in that period. ⁵³² Roman legions certainly did not exist in Palestine during this period, but if he means that no gentile auxiliaries were stationed there (under Agrippa’s authority), his claim would be remarkable, for such a supposition flies in the face of explicit contrary evidence from Josephus, a military writer who was alive and present in the period described.
Three responses seem appropriate: first, does the Cornelius narrative occur during Agrippa’s reign? (It may, but need not, have done so.) Second, what was the military presence in Caesarea at the time? Josephus explicitly refers to auxiliaries of the Roman army in Caesarea during this period. Third, some argue that Cornelius was retired anyway. Of these arguments, the most important is the second from Josephus. Ultimately, Luke’s report of Roman soldiers in Caesarea is not historically problematic. Indeed, it is the one place in Judea where one would expect to find them, apart from the cohort in Jerusalem.
During Agrippa’s Rule?
Agrippa’s rule is mentioned in Acts 12:1–23, so it appears that Luke’s story does not follow Agrippa’s death. But this conclusion need not invariably follow (probable as it is); Luke could have moved a later story earlier in his narration⁵³³ (so that Peter’s gentile mission, and Jerusalem’s approval, precedes the gentile ministry of the anonymous believers of 11:20). Peter’s “tour” of coastal cities (starting in 9:32) occurs only after his return to Jerusalem (8:25), and in 12:17 he left for “another place.” Thus some suggest that “the cohors Italica may have come in with the reconstitution of the province in 44,” and the story may be dated to that time.⁵³⁴
Yet the gentile mission should have been well under way in Antioch by then, and not merely beginning; thus another alternative seems likelier, and is plausible given a Pauline chronology that is (in my opinion appropriately) lower than usual: the events here might occur before Agrippa’s rule began in 41 CE.⁵³⁵ Nothing in Luke’s narration would preclude this option even if his sequence were strictly chronological here; Agrippa does not appear before Acts 12:1, and Luke offers no indication of the chronological distance between the present events and those of Acts 12.
In either case, we do not need to believe that Luke’s narration is strictly chronological here. Chronology was important to history and, where possible, to some other genres.⁵³⁶ Sometimes, however, it was not possible or desirable. Military historians like Polybius and Thucydides could provide careful chronology because they or their sources had annals or notes available;⁵³⁷ the same was not true when one’s sources depended on oral recollections.⁵³⁸ Thus, for example, Seneca the Elder promises to recall events of the past so long as he need not do so in sequence;⁵³⁹ Papias reports that Mark recounted Peter’s reports, but not in order.⁵⁴⁰ Biographies (like Luke’s first volume) were even less constrained by chronology than history was;⁵⁴¹ having surveyed Augustus’s life, Suetonius promises, “I shall now take up its various phases one by one, not in chronological order, but by classes, to make the account clearer and more intelligible.”⁵⁴² Even in historiography, when using episodic narrative “woven . . . around the impact of a personality,” a topical arrangement was appropriate.⁵⁴³
Sometimes historiography also required one to backtrack or advance into the future, to keep to a train of thought or geographical region. Thus Josephus sometimes follows a topic explicitly out of chronological sequence (Josephus Ant. 18.194); at other times he does so without explicitly informing readers.⁵⁴⁴ Dionysius of Halicarnassus complains that Thucydides’s chronological style breaks up his flow of thought and obscures his narrative, whereas Herodotus more naturally follows the flow of events.⁵⁴⁵ Only if one specified chronology (and in the process contradicted other historical sources) did one open oneself for criticism.⁵⁴⁶
Acts generally develops chronologically, but there are likely exceptions. For example, because Luke follows Peter from 9:32—11:18, then returns to the Hellenist summary of 8:1 in 11:19, his arrangement in 11:19–30 is as much topical (and somewhat geographic) as chronological (a tension all ancient historians had to straddle). Because Luke lacks clear chronological markers at this point, however (except that this evangelism began between Stephen’s death and Agrippa I’s persecution), we cannot say whether disciples in Antioch, like Philip, may have preceded Peter’s ministry to Cornelius’s household.⁵⁴⁷ For Luke, it was important that theologically Peter’s activity took precedence, so this may be why the Cornelius story appears before this summary.⁵⁴⁸ Because 11:19–30 interrupts Petrine material in 9:32—12:24, yet remains sketchy, Luke may have different sources and not know their relative chronology. While I am inclined to think that the Cornelius narrative historically preceded the movement in Antioch, my point is that clear chronology is lacking.
Yet even if we assume neither alternative, deciding that the narrative occurs during Agrippa’s reign, and even if that meant that he had his own Jewish soldiers in Caesarea, the area would retain many former soldiers who were from the area or who had settled there. We know that gentile soldiers lived in Caesarea then as they did afterward.
Soldiers in Caesarea
Literary evidence indicates that in most of the early first century Caesarea held a cavalry unit and five infantry cohorts of auxiliaries; archaeological evidence also attests an increasing Roman presence and influence.⁵⁴⁹
Soldiers were certainly stationed there after this period. During the later governorship of Felix, Jews provoked Syrians in Caesarea (Josephus Ant. 20.175). The Syrians had less wealth than Jews there, but counted on connections with the soldiers stationed there, most of whom were from either Caesarea or Sebaste (20.176). Conflict escalated, and when the Jewish side refused to desist, Felix set his soldiers against the crowd, and let them plunder the homes of well-to-do Jewish citizens (20.177). In times of conflict the Roman auxiliaries in Caesarea, who were mainly Syrian, would side with the local Syrians against the Jews (War 2.266–68).⁵⁵⁰ From this we may gather that, within a few decades of Cornelius’s encounter with Peter, the majority of the gentile soldiers in Caesarea were anti-Jewish, or at least would prove pro-Syrian if a conflict between the two groups arose. They also had local ties.
Most relevantly, an explicit passage in Josephus describes gentile soldiers already stationed in Caesarea at the close of Agrippa’s rule.⁵⁵¹ Many soldiers in Caesarea and Sebaste during Agrippa’s time hated Agrippa; they were nearly relocated after his death, but were allowed to remain, being very attached to the locale. Caesarea and Sebaste ungratefully rejoiced at Agrippa I’s death (Josephus Ant. 19.356); the numerous soldiers there went to his house and carried out images of his daughters, which they then abused on top of brothel-houses (Ant. 19.357).
The emperor Claudius was angry about this report (19.361), and ordered Fadus to go there and punish Caesarea and Sebaste for their insults both to the deceased Agrippa and to his living daughters (19.363). He then ordered that the soldiers in Caesarea and Sebaste be removed to Pontus, being replaced by Roman legionaries stationed in Syria (19.364). Nevertheless, these soldiers managed to persuade the emperor to allow them to stay; Josephus concludes that these same soldiers later led to disasters for his people (19.365).⁵⁵²
As Josephus’s narrative suggests, soldiers often became attached to local areas. By the second century, soldiers were usually recruited locally and stationed in a single location;⁵⁵³ such stability was less characteristic in the first century but more common than in earlier times. Military camps were often closely connected with their locales. In peacetime, soldiers could function as builders and engineers. Some of these projects were solely for their own benefit⁵⁵⁴ but others (like bridges and roads) would ultimately generate more civilian use.⁵⁵⁵ Rome also used military engineers and soldiers to help local public projects.⁵⁵⁶ Localities helped supply army camps, in turn helping to stimulate local economies.⁵⁵⁷ Temporary settlements (canabae) often grew around army camps, including merchants, artisans, and concubines who often bore children to the soldiers.⁵⁵⁸ It appears that the soldiers stationed in Caesarea felt at home there and wished to remain there.
It is virtually impossible lexically to deny that Josephus refers to soldiers in Caesarea during this time.⁵⁵⁹ Even if we were to argue that the anger of these soldiers stemmed from Agrippa removing them from active service (which the text does not indicate),⁵⁶⁰ they would be soldiers nonetheless (we cannot allow Josephus to employ that language and then forbid it to Luke).⁵⁶¹
Cornelius’s local connections
From Josephus’s account we learn that the soldiers stationed in Caesarea felt very attached to the locale; although the climate was pleasant, we may surmise that many were recruits from this area, or had entered into unions with local women. Granted, these relationships were not officially legal, but they connected soldiers to local areas and encouraged many to retire there. Roman law from Augustus onward prohibited soldiers from marrying, and Roman camps lacked domiciles for wives; any children born to their unions with local concubines were thus counted illegitimate and not Roman citizens.⁵⁶² It is even possible that any marriage that did exist at enlistment was dissolved.⁵⁶³ The military did not need to take local attachments into account when transferring troops, although this transfer became less common in later times.⁵⁶⁴
Nevertheless, avoiding local ties was difficult in practice. In this period soldiers normally served a minimum of twenty years, often from roughly ages 17 to 37. (The period was later extended to 25 years under Hadrian.)⁵⁶⁵ Centurions sometimes chose to stay longer, as career soldiers.⁵⁶⁶ Granted, because centurions were “often transferred from one legion to another,”⁵⁶⁷ forming long-term local relationships might be more difficult for them than for others.
Official policy against local marriages was rarely enforced. Soldiers formed marital unions with provincials, which the women often found economically suitable.⁵⁶⁸ Such unions could include “dowries from their wives’ parents.”⁵⁶⁹ Many soldiers probably considered themselves married, regardless of official rules, especially as legions became increasingly stationary in later times.⁵⁷⁰ Tariff lists reveal that the authorities knew that soldiers and sailors had attachments to local women.⁵⁷¹ Concubinage was unofficially permitted, and even a general could set the same example (Suet. Vesp. 3).⁵⁷² Legionaries in Cologne married locals (Tac. Hist. 4.65).⁵⁷³
These unions lacked some securities that legal Roman marriage would have protected.⁵⁷⁴ Once they were discharged, however, their children were sometimes or perhaps even often retroactively legitimated as a reward to the soldiers, provided each soldier sought it for the children of no more than one concubine.⁵⁷⁵ Claudius made the same grant to soldiers who had served 25 years or more.⁵⁷⁶
Roman law later noted that emperors typically allowed veterans to legally marry foreigners, counting (subsequent) offspring as Romans (Gaius Inst. 1.57). But what of offspring born during one’s term of service in the provinces? Even in earlier times, generals sometimes pardoned soldiers who stole away from camp nocturnally because of romance.⁵⁷⁷ In the late second century, Severus granted soldiers “the right to live at home with their wives,” legally ratifying a widespread existing situation but further weakening traditional military discipline.⁵⁷⁸
Roman law did recognize concubinage, but the loss of rank incurred by the woman generally led men to seek concubines of lower rank.⁵⁷⁹ Concubinage as a substitute for marriage was common in the early empire,⁵⁸⁰ and perhaps especially dominant among slaves and freedpersons.⁵⁸¹ A man could legally hold only one concubine, however, and not concurrently with a wife.⁵⁸² Roman law recognized concubinage, provided one had only one (and also not be married at the same time).⁵⁸³ In these circumstances, the children were related to the mother but not legally related to father, though he could bequeath to them property provided it did not infringe on the rights of legitimate heirs.⁵⁸⁴ The offspring’s lack of claim on an estate sometimes made concubinage preferable to a second marriage for a man with heirs.⁵⁸⁵
Auxiliaries like Cornelius, many of whom served near their homes (see discussion above), may not have faced the same restrictions as legionaries. “Senior officers and auxilia” could live with their wives before the time of Severus.⁵⁸⁶ Auxiliaries were normally not Roman citizens before the time of their discharge, limiting the effects of Roman concerns about the children’s status, though Cornelius’s name might suggest that he was a Roman citizen.
How relevant all of this is to Cornelius’s “household” (Acts 10:2) is unclear, since Luke does not specify whom it includes (cf. the servants in 10:7).⁵⁸⁷ It does illustrate, however, that local ties were quite common. Thus some retired soldiers and local recruits would have likely chosen to remain on in Caesarea.
A retired soldier?
Even if the events occurred in 41–44 CE, and even if we (against the evidence of Josephus) deny the possibility of active gentile soldiers residing there in that period, Cornelius could be a retired soldier (who may have served in the region) who settled in Caesarea, as his own home (perhaps in Caesarea rather than an adjoining camp) and the presence of family might possibly suggest (see comments above). Romans had traditionally provided land grants for veterans (sometimes—especially in earlier times—in a colony; but sometimes on their own).⁵⁸⁸ Although land grants were not systematic, they continued until the second century CE.⁵⁸⁹ By this period cash gifts appear more common than land grants (at least for common soldiers); but soldiers in any case usually preferred to settle “near their last camp” rather than in a veteran colony.⁵⁹⁰ Centurions became members of the equestrian order when they retired, which could further explain Cornelius’s prominence here.⁵⁹¹ Although veterans could choose to remain as reservists⁵⁹² and Rome would recall veterans to war in emergency situations (like the civil wars of the first century BCE), they normally spent their retirement in peace (Cic. Phil. 5.16.44).
In view of his access to a soldier in 10:7, I am inclined to think that Luke does not envision Cornelius as retired, unless (as is possible) this soldier is also retired, or Cornelius hired him as a part-time worker; but one could argue that Luke’s sources reflect a retired centurion. Some have even argued, on the grounds that Archelaus’s and Agrippa’s auxiliaries both passed into Roman service (in 6 and 44 CE, respectively; cf. Josephus Ant. 19.364–65), that Cornelius belonged to an auxiliary unit later absorbed into Roman service and attached to the Italian cohort in Syria.⁵⁹³ (Nevertheless, if so, I would argue from Ant. 19.356–65 that these units already consisted of gentiles.)⁵⁹⁴
While on the whole I think it likelier that Cornelius remains in active duty, the possibility that he was retired would allow us to accept Luke’s report of a gentile soldier converted even if the events occurred during Agrippa’s reign and he used only Jewish soldiers. If Agrippa temporarily supplanted the local Syrian soldiers with his own Jewish soldiers, a semi-retired state would be possible. But in the final analysis, these approaches merely illustrate that Haenchen’s position cannot be supported from any angle. The explicit presence of auxiliaries in Caesarea even during the reign of Agrippa makes clear that there is no reason to question Luke’s claim. Whether the gentile was ethnically Roman (a matter Luke does not explicitly address, though the name may imply citizenship by whatever means it had been acquired) is a separate question, but gentile auxiliaries clearly lived in Caesarea in this period.⁵⁹⁵
Conclusion
None of the evidence available undercuts Luke’s portrait of a gentile centurion becoming a Christian in the early days of Christianity. Cornelius could have been retired; could have converted after or (more likely) before Agrippa’s reign; or, for that matter, could have been converted during his reign. Josephus is clear that gentile soldiers existed in Caesarea during the reign of Herod Agrippa I.