A misleading symmetry
The standard narrative of 49 BCE casts the war as Republic against tyranny: Pompey defending the senate, Caesar attacking it. The narrative was Cicero's at the time; it has been a great deal of the European tradition's ever since. It is half-true.
What it leaves out is the parallel structure of the two careers. By the time the war began, both Pompey and Caesar had built their political weight on the same set of late-Republican departures from constitutional norm — extraordinary commands voted outside the regular cursus, armies whose veterans' settlement depended on their general's continued political reach, alliances with other power-holders that operated outside the senate's formal procedures.
Pompey's career, in particular, is hard to defend as a model of constitutional restraint. He was awarded the cognomen Magnus before he had held any regular magistracy. The lex Gabinia of 67 gave him an extraordinary command against the pirates that was unprecedented in scale. The Mithridatic command followed under the lex Manilia. His Eastern settlement of 63 created Roman policy on a scale no senate had previously delegated. The sole consulship of 52, dictatorial in everything but name, was his.
The argument that the senate's eventual alignment with Pompey constituted a defence of the Republic depends, then, on a specific claim: that Pompey, despite the structure of his own career, would in the end consent to be constrained by the institutions in a way Caesar would not. The claim is plausible. It is not the obvious reading the senatorial faction took it to be.
What Pompey was and was not
Pompey was, by Roman standards, conservative in temperament. He had served under Sulla; he respected the old senatorial families even where his own career trespassed on the customs they preserved; he had walked away from his own opportunity for tyranny in 62 BCE, disbanding his army on his return from the East before entering Italy. Cicero's correspondence shows him repeatedly hoping — sometimes despairing — that Pompey would in the end choose constitutional order over personal advantage.
Whether the hope was justified is debated. The strongest evidence for it is precisely the disbandment of 62: Pompey could, in that moment, have done what Caesar would do eleven years later, and he did not. The strongest evidence against is the structure of his career, the unprecedented commands he had not refused, and the private nature of the Triumvirate.
Why the senate chose him
The senatorial faction's eventual support for Pompey against Caesar was not, in the senatorial faction's own reading, a decision between two careers it admired equally. It was a decision between two careers it could no longer treat as equivalent. Caesar's command in Gaul had run continuously for nine years; the standing precedents required him to give it up before standing for a second consulship; his proxies in Rome had blocked every senatorial attempt to obtain that result; his army's combat strength and political reach were both, by 50 BCE, larger than anything in the Roman political imagination outside Pompey himself. The senatorial faction calculated, not unreasonably, that the form Caesar's resumption of political life would take if he were allowed to come home with his army intact would not be one the Republic survived.
The calculation was correct about the destination. It was wrong about the alternative. The Pompeian war was not won by the Pompeian side. Pharsalus, in August 48, was a decisive defeat. Pompey fled to Egypt and was murdered on landing. The senatorial remnant under Cato fought on in Africa until Thapsus in 46 and in Spain until Munda in 45. None of it brought the Republic back.
What the long tradition has read
The long tradition has read the war in three principal ways. The senatorial reading (Cicero, Lucan, the early-modern republicans, the American founders) treats the Pompeian cause as the Republican cause even granted Pompey's career, because the alternative was worse and was known to be worse. The skeptical reading (parts of Plutarch, much of the imperial-era historians, much modern scholarship) treats the war as a contest between two careers that had each, by the standards the senate professed, already departed from constitutional norm — a fight over which kind of departure the political class would tolerate. The third reading — Tacitus' implicit one — treats the war as a contest whose outcome did not matter for the question of the Republic's survival, because the Republic had been substantially gone since Marius.
The platform does not need to choose. The question itself — what the institutional defence of a republic in late stages of decline can plausibly consist in — is what the war made visible. Its answer is uncomfortable in any direction.