Tacitus traces a series of conflicts between Agrippina the Elder and her father-in-law Tiberius. After the death of her husband Germanicus in Syria (a.d. 19), Agrippina returned to Rome with their children. Germanicus' lingering popularity with the armies and people meant that his widow Agrippina and their children enjoyed a level of popular support as well—one that eventually became a threat in Tiberius' mind. Agrippina, moreover, refused to embrace the modest, retiring role that her father-in-law (and Roman society in general) expected of her. Tiberius was, moreover, ‘never gentle toward the house of Germanicus’Footnote ² and his concerns were augmented by the machinations of Sejanus, who reported that the people were dividing themselves into factions, some even calling themselves members of the partes Agrippinae.Footnote ³ The combination of Agrippina's high birth—the only still-living grandchild of Augustus—and her status as widowed daughter-in-law of the emperor, therefore mother of the emperor's probable heir, along with her persistent independence and sometimes unfeminine strength of character made her seem an intolerable political threat. This paper examines the culmination of these conflicts, when Agrippina is subjected to criminal prosecution and penalty in a.d. 27 and 29 at the instigation of Sejanus, with the overt approval of Tiberius. Of primary concern is the timing and the order of the charges brought against Agrippina.

As a starting point, a review of Tacitus' account of Agrippina's demise is in order. Criminal proceedings directly against Agrippina began after her relationship with Tiberius had already deteriorated and Sejanus had successfully removed several of Agrippina's friends and supporters.Footnote ⁴ Tacitus presents an incomplete report of accusations against Agrippina as the last item in his account of the year 27:

This kind of accusation, one would think, should lead to very serious charges—no doubt of a treasonous flavour—particularly considering the threat of flight to and support from the armies in Germany.

Another incomplete presentation of accusations appears two years later under the year 29, immediately after Livia's death (which is recorded as the first item in that year's account). Tacitus reports that Tiberius wrote a denunciatory letter to the Senate from his secluded position at Capri:

The Senate listened in severely shocked silence (magno senatus pauore ac silentio).Footnote ⁷ At last Cotta Messalinus, consul of 20, came forward with a frightening proposal (cum atroci sententia), but the cautious and confused senators, many of whom must have supported Agrippina, delayed and did nothing.Footnote ⁸ Anonymously written pamphlets criticizing Sejanus were circulated in the city containing proposals allegedly made under the names of the consuls. Tiberius responded angrily that the Senate and people had spurned his instructions and insulted his agent and so, repeating the abuse against Agrippina and Nero, he said that he would decide the matter himself.Footnote ⁹ Tacitus' account breaks off at this point and there is a lacuna in the text that spans approximately two years.

The tangled chronology indicates that there are some problems with Tacitus' account. He placed the final attack against Agrippina early in the year 29, immediately following Livia's death.Footnote ¹⁰ Three other authors, however, indicate that there was a conviction before Livia's death. Suetonius records that Livia had cared for young Caligula after his mother was banished.Footnote ¹¹ Pliny implies that Agrippina's son Nero, who was accused at the same time as his mother, had been condemned before Livia's death; Velleius, too, indicates that Agrippina and Nero were convicted before the death of the emperor's mother.Footnote ¹² One way to reconcile Tacitus' account with those of Suetonius and others is, as Meise argues, to conclude that Tacitus has produced a corrupt and incomplete account of what were really two attacks: the first in 27 that resulted in house arrest, and a second and more serious one in 29 that resulted in the permanent exile and subsequent deaths by starvation of both defendants.Footnote ¹³

More is known about Agrippina's punishments than the legal process that gave rise to them, and this information can be used to deduce the accusations she most likely faced and to fill in some of the gaps left by Tacitus. Various ancient authors agree—with no specific dates—that she was exiled to Pandateria, an island off the coast near Naples.Footnote ¹⁴ When she and Nero were moved, they travelled in closed litters under heavy guard; Nero was even declared an enemy of the state.Footnote ¹⁵ Agrippina was a non-compliant prisoner—perfectly in keeping with her personality—and suffered beatings and force-feedings, the first ill-treatment recorded for an exile of the imperial family. Ultimately, however, her determination to starve herself to death was successful and she died on 18 October 33, two years to the day after Sejanus died.Footnote ¹⁶ Although she did not suffer the additional posthumous penalty of the destruction of her public images (what we might call damnatio memoriae), her birthday was added to the days of ill omen.Footnote ¹⁷ Tiberius boasted of his clemency because he did not strangle her and throw her body on the Gemonian Stair, a traditional punishment for someone convicted of the most serious treasonous offences.Footnote ¹⁸ The penalties Agrippina experienced were not ones assigned lightly, and were not penalties in keeping with the crimes of arrogant speech or even writing libelous pamphlets.

Agrippina's punishments are consistent with other cases in Julio-Claudian times when fifteen other women were charged with crimes against the emperor's person, his family or his position.Footnote ¹⁹ Like Agrippina, five of these women were punished with some form of exile. Agrippina's close friends Sosia Galla and Claudia Pulchra were in this small group as were Aemilia Lepida, Sancia and Lollia Paulina. Other women suffered capital punishment; Aelia Junilla along with other members of her family and Livilla (Livia Julia), guilty largely owing to their close association to Sejanus, were executed in the aftermath of Sejanus' downfall in 31. A further four women committed suicide because of the charges brought against them: Mutilia Prisca, Paxaea, Cornelia and Marcia Servilia. In the context of these other women's punishments, Agrippina's exile in 29 seems suitable for a conviction based on accusations of maiestas rather than a lesser charge.

One more piece of evidence remains: a chance comment preserved by Philo reveals that Tiberius' friend Avillius Flaccus was one of Agrippina's accusers, after which he was made prefect of Egypt.Footnote ²⁰ He served as prefect from about a.d. 32 until he was recalled to Rome by Caligula in 37; the timing seems most closely connected with the attack on Agrippina in 29, but what charges, exactly, Flaccus presented against Agrippina are unspecified. Although Tacitus has recorded Tiberius' invective against Agrippina, purportedly contained in the letter sent immediately after Livia's death, there were no viable legal accusations in the recorded communication. The inflammatory pamphlets mentioned as circulating in 29 are not directly attributed to Agrippina or Nero in our sources, but could have theoretically formed the basis for some kind of libel accusation. Tiberius had previously refused to allow prosecution for verbal insults, but written attacks—especially anonymous ones—were taken more seriously.Footnote ²¹ None the less, if Avillius Flaccus received the post of prefect of Egypt as a reward for his role in accusing Agrippina, it would have been a more appropriate reward for a successfully concluded high-treason case rather than for a case based on her undisciplined speech, or even one based on libelous pamphlets.

Although scholars and ancient authors agree that the second attack in 29 was more serious and resulted in the more serious punishment of exile to Pandateria, Tacitus records the more serious accusations in connection with the earlier charges in 27. There were ‘soldiers who reported Agrippina's every move’ in Tacitus' account of 27, strongly suggesting that Agrippina and Nero were under some sort of guard, likely at the villa in Herculaneum indicated by Seneca.Footnote ²² But why were they under guard or house arrest in the first place? In fact, Tacitus indicates that the very soldiers assigned to guard them collected the information that led to the accusations which resulted in their initial arrest. Something is very wrong in this scenario. Moreover, why did the Senate react with shocked silence to a letter in 29 describing the (relatively trivial and rhetorically timeworn) moral offences of two already imprisoned convicts?Footnote ²³ This is a reaction more in keeping with an initial and unexpected accusation than one that followed a conviction and two years of house arrest. Perhaps most insoluble, why did very serious charges (even if completely fabricated), based on a genuine connection with the restless Rhine legions, result in house arrest at a grand villa in Herculaneum in 29, but accusations of arrogant speech and truculent character in 27 resulted in exile to Pandateria, an already notorious prison island?Footnote ²⁴

This historical problem appears a veritable Gordian knot at first assessment; but there may be a solution, and one that is very simple. If the contents of the two sets of accusations are reversed, chronological problems and contradictory source problems, as well as Tacitus' internal inconsistencies, disappear. Now we can see a more logical escalation of action against Agrippina. Following the proposed revision, in 27 Agrippina and Nero were denounced for moral transgressions in a letter written by Tiberius. Shocked and confused, the Senate did not know what to do with the legally inactionable charges and delayed, while public demonstrations claimed the letter was fraudulent. Tiberius responded sharply that he would deal with the matter himself. Agrippina and Nero were placed under house arrest with soldiers assigned to guard and spy on them. The soldiers sent their reports to Sejanus, who gathered (or suborned) enough damning information about talk of flight to the German legions and appeals to the Senate and people, while claiming sanctuary at the statue of the deified Augustus to launch the second attack in 29, this time on charges of (probably) maiestas. Agrippina and Nero were convicted; the latter was declared a public enemy.Footnote ²⁵ They were punished with relegatio and sent to Pandateria and Pontias, respectively, where they eventually died. When they were moved from Herculaneum to their final places of exile, they travelled in closed litters with a military escort to prevent contact with anyone they might meet on the road,Footnote ²⁶ a measure more in keeping with preventing traitors from contacting supporters than with protecting a jaded public from views of the morally reprehensible. In further support of this revised scenario is a reference in Suetonius that links the charges of treason (listed by Tacitus in the earlier set of accusations) with Agrippina's exile to Pandateria: nouissime calumniatus modo ad statuam Augusti modo ad exercitus confugere uelle, Pandateriam relegauit, ‘at last he relegated her to Pandateria, falsely charging that she wished to flee to the statue of the deified Augustus or to the armies’.Footnote ²⁷ Suetonius' record connects the more severe penalty with the more serious charges.

This problem in Tacitus' account is not a textual problem in the Annales that could be solved by a clever modern editor or blamed on textual transmission, but simply an error. While he is widely acknowledged as the overall best narrative source for the Julio-Claudian period, and is lauded for insights and analysis not found elsewhere, Tacitus' account is not infallible, particularly in regard to details of chronology.Footnote ²⁸ There are some clues, however, about how he might have introduced such an error. The record of Livia's death in 29 is a temporal anchoring device in his narrative: she died, and there was a letter, followed by the final destruction of Agrippina and Nero. The heart of the error is Tacitus' report of the contents of that letter. By late 26, Tiberius had retired from Rome and all of his communication with the Senate and other officials in Rome was conducted by letter, so there is no obstacle to moving the date of the ‘arrogant-speech’ letter two years earlier, since nothing external holds this particular letter to the year 29.Footnote ²⁹ Tacitus is very clear that a letter to the Senate was made public immediately following Livia's death, but since all of Tiberius' communication with the Senate was conducted by letter, this is hardly definitive proof of the contents of the missive. He simply reported the contents of a letter sent at an earlier time. Tacitus himself even records the rumours that the public believed that the letter had been received rather earlier than it was read out, but had been withheld by Livia and was only released after her death, or that the letter was an outright forgery.Footnote ³⁰ The nature of Tacitus' error suggests that he inaccurately compiled his narrative from two different accounts, suggesting that he made less use of primary records such as senatorial minutes than we might wish.

To sum up: a comparison of Tacitus' account with references in Suetonius, Pliny and Velleius confirms that Agrippina faced two sets of charges, first in a.d. 27 and then in 29. Even when Tacitus' account has been reinterpreted in light of a second set of accusations against Agrippina, I argue that he has inverted the order of the charges, describing the very serious charges as part of the first attack, and the trivial charges as the basis for the second attack. When the order is reversed, then a more logical escalation of action against Agrippina is seen, beginning with arrogant speech and insulting behaviour and progressing to a treasonous threat of fleeing to the army. The accusation that Agrippina was planning to exploit her connection to the German legions and launch a civil war was taken very seriously, so her conviction is unsurprising, but the physical abuse of Germanicus' widow and the emperor's own daughter-in-law while she was in captivity is unprecedented. The Romans' lingering anger against Agrippina might be explained if they felt betrayed because the emperor's daughter-in-law was convicted of taking action to generate a civil war, and, even more importantly, because they believed she had the power and influence to actually do it.