Every friendship looks solid the day it’s formed. The only test that actually means anything is what it looks like twenty-five years later, after at least one year in the middle where it nearly wasn’t.
In 58 BC, Cicero was forced out of Rome — exiled, his house on the Palatine burned, his political career apparently finished — and in the letters he wrote during that miserable year, something unusual surfaces. He blames his friends. Specifically, he suggests that Atticus, the man he would later call closer to him than his own brother, had counselled caution when boldness might have saved him: “Even Atticus,” one biographer’s summary of the correspondence puts it, “he thinks was timid, at the best, in advising his retirement.”
It is, by every account of the twenty-five years of letters between them, the one moment where a crack shows.
This is where Part V of this series has to start, because it is the detail every tidier account of Cicero and Atticus’s friendship prefers to skip. The two men had been close since boyhood. They stayed close for the rest of Cicero’s life. But “stayed close” did not mean the friendship never strained — it meant the strain, when it came, did not end anything. That distinction is the entire subject of this final part.
The friend who left for Athens
Titus Pomponius — the man history remembers by his adopted name, Atticus, for his decades in Athens — made an unusual choice early in his adult life. While Cicero pursued the Roman bar and Roman politics with everything he had, Atticus left Rome after his father’s death and spent roughly twenty years in Greece, studying philosophy, managing his considerable fortune, and deliberately keeping himself out of the factional violence that was already tearing the Republic apart.
This was not indifference to Rome, and it was not cowardice, though both readings are available if you want them. It was closer to a structural decision, in the same spirit as Part IV’s argument about brotherhood requiring design rather than luck: Atticus built a position for himself — geographically removed, financially independent, formally unaligned with any faction — that let him remain in genuine contact with men on every side of Rome’s widening civil conflicts, including men who would shortly be trying to kill each other. He could write to Cicero and, separately and just as warmly, maintain relations with men Cicero considered dangerous, without either relationship compromising the other.
This is worth naming precisely, because it is not the same as simply avoiding conflict. Atticus stayed engaged with everyone. What he refused was to let any single alignment become the thing that defined him, because he correctly foresaw that Rome’s politics would keep changing who was safe to know and who wasn’t, and that a friendship built on shared faction would not survive the next reversal. A friendship built on something else might.
The crack that didn’t become a break
The exile episode matters precisely because it complicates the tidy version of this story. Cicero, humiliated and displaced, needed someone to blame beyond himself, and for a period the letters show real resentment directed at the friend who should have been unimpeachable. This is not a minor detail to smooth over in service of a cleaner narrative about ancient loyalty. It is the whole point.
Every friendship that lasts decades will, at some stage, pass through a year like this — a period where one party is convinced, rightly or wrongly, that the other failed them at the exact moment it mattered most. The friendships that don’t survive are not necessarily the ones where this happens; it happens in most of them eventually. The ones that survive are the ones where neither party mistook the bad year for the final verdict on the whole relationship. Cicero’s resentment during his exile did not end the correspondence. It appears, reads bitterly for a stretch, and then the letters continue much as they had before — hundreds more of them, across the following decade, on every subject from politics to real estate to the trouble with Cicero’s brother’s marriage.
Four hundred and fifty-four letters
The sheer volume is worth sitting with as its own kind of evidence. Of Cicero’s 813 surviving letters, 454 are addressed to Atticus — written across roughly twenty-five years, from 68 BC to Cicero’s death in 43 BC, through consulships, exile, civil war, dictatorship, and the final collapse of the Republic both men had spent their adult lives inside. The Roman biographer Cornelius Nepos, who knew them both, put the relationship in the plainest terms available to him: “Marcus Cicero loved Atticus above all men, so that not even his brother Quintus was dearer or more closely united to him.”
Twenty-five years is not a figure of speech. It is longer than most modern friendships get tested at all, let alone tested against the collapse of the political system both men had built their lives around. The letters are not uniformly warm — they complain, they nag about money, they relitigate old grievances about Atticus’s brother-in-law relationship to Cicero’s family, they occasionally go quiet for stretches neither man explains. What they do not do, across a quarter of a century of exactly the kind of material that ends most friendships, is stop.
What was left to write at the end
In 44 BC, with Caesar recently dead and Rome sliding toward the chaos that would eventually kill him, Cicero — in his sixties, largely excluded from the power he’d spent his life pursuing — sat down and wrote De Amicitia, an entire treatise on the nature and value of friendship, and dedicated it to Atticus.
By that point there was very little left that Atticus could do for Cicero in any practical sense. Cicero was not seeking patronage, or protection, or a favour that required calling in decades of goodwill. He was, near the end of a life that had run through every kind of triumph and humiliation Roman public life could offer, choosing to spend some of his remaining time writing down what the friendship had actually been worth — not as a debt being settled, but as an account being kept honestly, the way Seneca kept his own account each night, for a relationship rather than a single day.
Cicero would be dead within the year, killed on the orders of Mark Antony. The letters to Atticus continue almost to the end.
What the long game actually asks
Part II of this series described what Aristotle thought friendship required: living together, shared time, the truest form reserved for good men who’d proven it to each other across years. Part III named the gap that stays open between men who want the same closeness but won’t do the work to close it. Part IV argued that closing it takes structure — brotherhood built on purpose, not left to hope.
Part V adds the piece none of the others could supply on their own: none of it is tested until decades have actually passed, including at least one stretch where it looked, from the inside, like it might not survive. Cicero and Atticus were not friends because nothing ever went wrong between them. They were friends because a year that could have ended it didn’t, and twenty-four more years followed it, and at the end there was still something worth writing down.
That is the long game. Not the absence of strain. The twenty-five years on the other side of it.
Brotherhood — The Friendship Problem, a five-part series on why men lose each other after thirty and what actually closes the distance. This concludes the series.
