IF the Arabian (Nabatean) King Aretas ever occupied Damascus, it would have been before 37 AD. It could not have been after. Ogg missed this. Jewett missed this. Bowersock pointed it out in 1983 and few have acknowledged it since. The historical context is vital to Pauline chronology AND to the chronology of the earliest church in Jerusalem.
Here's the very-skinny. In 20 BC, the Kingdom of Zenodorus was granted to Herod the Great even though it had been promised to Nabatea. The Nabateans made trouble in Trachonitis until Aretas betrothed his daughter to Antipas (c.1 BC/1 AD) and Philip managed to forge good relations with the Nabateans in his Tetrarchy. But Antipas broke the treaty when he married Herodias (28/29 AD) and Philip's death (33/34 AD) filled the old Kingdom of Zenodorus with an absolute power vacuum.
Tiberius (undoubtedly with, through or by proxy of Macro, the new Praetorian Prefect after Sejanus) officially annexed Philip's Tetrarchy into Provincia Syria. But Syria had been suffering from a power vacuum of its own. The Proconsul Lamia was an absentee Governor for ten years until Pomponius Flaccus [not to be confused with the Egyptian Prefect hated by Philo] arrived in 32. But Flaccus died in office in 33 and Tiberius (and/or Macro) sent L. Vitellius in 35, more than a whole year after Philip and Flaccus had both died.
Presumably, Vitellius was to establish the new status of Philip's Tetrarchy, but Vitellius had his hands full immediately with conflict on all sides. Dealing with the Parthian invasion of Armenia occupied Vitellius' first two summers while the Governor also sent one of his four Legions to help Cappadocia against a mountain tribe of Cilicians. Meanwhile, Herod Antipas had taken the liberty of sending his own small army to occupy the strategic fortress-city of Gamala in the Golan Heights. But while Antipas was at the Euphrates making peace with the Parthian King Artabanus, the Nabatean army took Gamala and crushed Herod's army.
By early 37, Vitellius was marching south, but purposely dawdled, resenting Antipas for taking credit about the Euphrates treaty in a letter to the Emperor. Tiberius (and/or Macro) had ordered Vitellius to avenge Antipas, but Vitellius lingered in Jerusalem after Passover until news arrived of Tiberius' death. At that, the Governor took his Legions back north. Gamala had already been reclaimed (officially for Syria) and Aretas had long since retreated. And just by coincidence, almost simultaneously, in Rome, the new Emperor Caligula (and his chief advisor, Macro) were appointing Antipas' nephew, Herod Agrippa, as the new King of the old Kingdom/Tetrarchy.
According to our records, Aretas did not attack or press through Trachonitis under Agrippa. It is extremely doubtful that Aretas could have managed possessions from the other side of Agrippa. And Aretas was somewhere in his 60's already, at least. He had been king since 9/8 BC. Two years after Caligula made Agrippa King of Trachonitis and the Golan, Aretas died, in 39 AD.
That's the whole skinny. Now here's the point.
It had long been assumed, by a very poor reading of 2nd Corinthians 11:32, that Aretas must have been granted Damascus by Rome, and the next argument went that since Tiberius sent Vitellius after Aretas, it must have been the nutsy Caligula. These arguments required skepticism of Josephus on Gamala as the point of battle, since the Golan was not an official "boundary" between Antipas and Aretas. But Josephus said Gamala, so the territorial issues must go back to the old grudge over Zenodorus. Only Bowersock (Roman Arabia, 1983) makes complete sense out of Tacitus, Josephus and Paul on this issue.
My own tiny contribution to this conversation is that Macro alone should be enough to debunk the old argument that Caligula suddenly did an about face from Tiberian policy. For all practical purposes, Macro was running the Empire in all twelve months of 37 AD, besides which Caligula never showed any interest in foriegn policy, except for the Temple worship fiasco in 39/40. Caligula merely gave his 'uncle Herod' a Kingdom as a reward for his friendship in recent years. Herod's Kingship, of course, is another issue the old arguments failed to deal with. If Caligula had wanted to give Damascus to anyone, it should have been Agrippa.
The Conclusion: Paul's "three years" in Arabia must end before winter of 36/37 and therefore his conversion must be dated to 33/34.
The Challenge: If we also take 33 as the year of Christ's Passion and Pentecost, what does that do to our view of the earliest church in Jerusalem? Tentatively clinging to 30 AD, which has become increasingly difficult to defend in recent decades except by appeal to tradition, seems to be motivated in some cases by a bias towards keeping Acts 1-8 in a long stretch of years. I think it was less than four months, but that's a story for some other time...