Small things change the course of history. A large event though prompted a decision in the history of Antioch that might very well have been the fatal flaw that sealed the city's fate and cast it into 1,500 years of obscurity.
In 526 AD a massive earthquake devastated the city but most severely damaged was the Island. The colonnades were thrown down, the palace damaged and the Domus Aurea was nearly toppled.
The upshot of this damage was that, seemingly, a decision was made to abandon the Island. The area was used as a quarry for stones to build the Wall of Justinian around the main part of the city. The Golden Octagon lingered on (now outside the city walls) until it eventually collapsed later in the century.
What if, however, the Island had not been the worst hit part of the city? What if the decision was made to fortify the Island instead of the mountainous part of the city? What might have been an alternative outcome?
The Island was the easiest part of the city to defend. If a strategy to strengthen the Island and use it as the fortress in time of extremis have been adopted then there might have been a chance that the city may have more effectively resisted the onslaught of Chosroes in 540 AD, and devastation from which the whole city went into terminal decline. As the reports of the seige make clear the Sassanid invader gained access to the city via the walls on the southern heights. These were the walls that had only just been built.
Frankly despite the enormous effort of constructing the Justinian Walls, the city was essentially indefensible in the new layout, it was too extended and the high parts of the city were too vulnerable. These encompassed, quite literally, two mountains. Defenders could not easily race to the walls.
With time the branch of the river that separated the Island from the main city became silted up, thus making the Orontes less of a moat and more of an access point. While still swampy the Crusaders in 1098 found the flatlands on the river-side of the city the most suitable place to make camp and sally forth against the Dog Gate and the Duke's Gate.
Why was the Island not chosen as the safe haven for the city? One earthquake decided its fate? We suspect though that the factors were twofold. Firstly, the Island was a sector of villas, baths and palaces. Maybe it also had stables (associated with the hippodrome) and garden areas. This implies low density. This also implies the bulk of the population that needed defending were in the other parts of the city. Thus the focus fell there with a priority of defending the majority, even if the task was well-nigh impossible. The imperial palace had not been used much by the emperors as a residence since the days of Valens over 140 years beforehand.
Then second factor is more prosaic. We have dwelt elsewhere on the possibility that the Island was supplied with its water from an aqueduct that came from the mountains to the north of the city rather than from the system geared to the springs at Daphne that supplied the rest of the city. While the Daphnetic aqueducts contained long surface and underground sections, an aqueduct from the northern mountains would have been elevated across the plain. It also could have been a prime victim of any earthquake. While not mentioned anywhere, it could very well be the case that the Island was left waterless by the earthquake of 526 AD and thus it was a "no-brainer" to rebuild the "mainland" part of the city and cede the Island to the limeburners and scavengers of stone and bricks.
Of course the builders of the Wall of Justinian did not imagine that plague, earthquakes and wars would reduce the city of over 500,000 in the 6th century to little more than 50,000 by the time the Crusaders appeared. The latter number would have easily fitted in a well-defended fortified Island instead of being sprawled across the existing fortified area with its gardens and farms because the walled area by the 11th century was way too large for practical defense.
When one considers the massive task of demolishing the Island walls, all its remnant buildings (including the hippodrome) and redeploying all this material into the construction of the city walls (so strong they lasted until the mid-19th century), then rebuilding the Island and creating a fortified city on the Island (and part of the mainland) and an unfortified portion for the rest (which seemingly was rather unfortified as evidence for the walls of Tiberius being of much substance are scance) would seem in retrospect to have been a more sensible strategy.
The road not taken by Justinian was to defend the defensible (i.e. the Island) and instead a grandiose wall-building campaign ultimately left the city more vulnerable, not less.