Recently, buried among a number of maritime news items, there was a two-sentence entry noting the anniversary of an “event” at sea. This provoked a flashback and the reminder that the day of this event 53 years ago marked not only a collision, but the end of an era. This newsletter then, is a mixture of the Rules of the Road and of a memory.
That evening I was the duty officer aboard the destroyer USS Monssen (DD-798) moored outboard of USS Cassin Young (DD-793) at the destroyer pier in Newport, R.I. Sometime after midnight the officer on watch came with an emergency message — our Destroyer Division was to get underway immediately. The engineering officer reported steam up on two boilers, ready to go on line and was getting steam up on the remaining two. The other two DDs of our division were astern of us and activity aboard all four had suddenly erupted as crews ran down the dock heading to all four ships. The commanding officers had been called by the destroyer force command, so they were already on the way.
We pivoted slightly on an after bow-spring, then backed away from Cassin Young. As we took Dumplings to starboard and then Castle Hill to port, the chief engineer reported three boilers on the line; and by the time Beavertail Light was astern to starboard, we had all four. The division formed a column with all four making turns for 25 knots, working up to 30.
Andrea Doria was sinking.
Given the name of a famous 16th century Genoese admiral, the 29,100-gross ton, 697-foot passenger liner Andrea Doria was the pride of the Italian maritime fleet. A year or so before, while operating with the Sixth Fleet, we had seen Doria in the Med — thus in some way we “knew” her. Her designer had captured something that has long since disappeared from the oceans. At rest she seemed straining to be unleashed — in motion she was grace. Single-stacked with a gentle sheer, she was meant to be at sea.
And so she was that July evening in 1956, steaming in fog in the westbound traffic lane en route to New York. The lanes between Nantucket and Ambrose lightships, with their 20-mile separations, were then advisory only — agreements between shipping companies, but not between nations. Nearing the end of her crossing, her fuel tanks were low, but financial/time constraints militated against seawater ballasting. Although her stability characteristics precluded a safe list beyond 20° (that would permit overflow of watertight bulkheads), such a list was considered an unlikely event.
Passing Nantucket a mile to starboard, she was making 23 knots on 268 T for Ambrose, 180 miles ahead. The master and two officers were on the bridge, the watertight doors were secured, fog signals were being sounded and speed reduced slightly to 21.8 knots. About 25 minutes after passing Nantucket, a radar echo 4° to starboard at 17 miles appeared on the screen. Subsequent radar bearings suggested a slow drift right with an estimated closest point of approach (CPA) one mile to starboard. One disquieting fact was the 40-knot speed of closure suggesting an eastbound vessel in the westbound lane.
Stockholm, a 11,700-gross ton, 525-foot passenger vessel of Swedish registry, had left New York that day, intending to take Nantucket one mile to port en route to Scandinavia. The third mate had the watch with orders to maintain at least a one-mile CPA on any contact. She was on course 087 T at 18 knots in the westbound lane on a moonlit night, calm sea state and good visibility. A northerly set prompted two small course changes — to 089 T then 091 T to compensate. (It was noted that the helmsman on duty had a tendency to yaw the vessel.) The watch officer noted a radar contact at 12 miles and slightly to port, but no visual contact. Subsequent bearings yielded an estimated CPA of one-half to three-quarters of a mile to port, inside the master’s night orders for a minimum CPA, but the mate delayed a course change pending visual sighting. The radar bearings continued a slow drift left as the range closed to four, then three miles — still no sighting. (The unstabilized radar was providing relative bearings — a “heads-up display.”) The helmsman’s noted tendency to yaw raised questions as to those relative-bearing accuracies when measured against instantaneous headings. Speed remained unchanged.
Meanwhile aboard Doria there was mounting concern over an unseen vessel with a high closure rate sounding no fog signal despite heavy fog. Conflicting instincts — the old law of the port helm, long discarded for all except “vessels approaching end on or nearly end on;” the inbred aversion to the idea of crossing an approaching vessel’s bow or projected track vs. the “customary practice of seamen” prescribing a port-to-port passing when meeting; the somewhat cursory recognition of radar given by the ColRegs of that day. All of these were lurking in the background of a decision in what is arguably the most dangerous of approaches in good or restricted visibility — a meeting fine on the starboard bow.
An unseen and unheard vessel with a high closure rate was approaching Doria with its CPA moving slowly starboard. The contact closed to two miles without visual sighting, but still no fog signal was heard. With speed remaining as before, the master ordered a 4° course change, left to 264 T, to open the CPA further to starboard. The first step in what came to be known in fine-meeting situations as “the waltz of death” had been taken — one vessel turning port, the other starboard.
At one and a half miles (by radar) Doria‘s lookout reported lights approximately 30° to starboard. Two mastheads opened to the right and a faint green side light confirmed the initial impression of a starboard/starboard passage. Suddenly, the aspect of the lights started to change. The lower masthead swung beneath and to the left of the after light, the green faded and then brightened to red. What initially seemed to be a close green-to-green passing had suddenly changed to a situation in-extremis — the master ordered hard left (with two short blasts).
Aboard Stockholm during those last three minutes, at just short of two miles, her lookout reported lights approximately 25° to port with mastheads open to the left and showing a weak (what was thought to be) red, supporting the initial estimation of a port-to-port passage. Step two of the macabre “waltz” had been taken when a 20° course change to starboard was ordered to widen the CPA. Unexpectedly, but quickly, the mastheads on the approaching vessel closed, and it was now showing green. Stockholm‘s watch officer ordered hard right and engines back full.
The rest is history. Five years and a month after she was launched, Andrea Doria, lanced deeply to starboard by Stockholm, went to the bottom. The collision took 51 lives, five aboard Stockholm and 46 aboard Doria.
Out-of-court settlements precluded resolution as to cause, although theories and opinions abound. It’s generally agreed that the perceived aspects of approach, as seen and recalled by the two bridges were mutually exclusive — if one was correct, the other had to be wrong. It seems to be generally accepted that the analysis of Capt. Richard Cahill in his Collisions and Their Causes, third edition, comes closest to the facts — that Stockholm commenced her fatal turn to starboard BEFORE sighting Doria visually. Contributing in part to her impulse to act were the mistaken relative bearings of Doria due to a combination of radar error and yaw — putting Doria (in reality) slightly to starboard of Stockholm before the latter’s turn rather than to port. Radar had guided their decisions into the jaws of collision.
So what comes out of this tragedy? Two tendencies which persist to this day in both clear and restricted visibility are failure to take early and significant action to avoid even the risk of collision and the acceptance of an unnecessarily close CPA. Repeatedly, court decisions have drawn attention to the idea that even if the initial maneuver was a poor decision, if done early, it would “provide time to maneuver out of a bad situation.”
About 25 miles from the scene, we received a message that we were detached; Ile de France and several other vessels were on the scene and the passengers were safe. As our column came around, we slowed briefly, 90° from the direction where she went down. Many aboard had seen her in the Med, and our ship seemed very quiet as we steamed back to Newport. With one exception, our division is long gone now. Cassin Young is currently a museum ship at the old Charlestown Navy Yard across the pier from USS Constitution.
Stockholm has gone through several owners, names and flag and she currently wears the name MV Athena. She will be 62 years young in 2010, and due to age, mounting safety requirements, etc., she may well end her days that year — ironically flying the flag of her current registry — Italy.
As for Andrea Doria, she left with us her image — that of a “ship.” Stately in design with grace in motion, she and those like her have gone forever, replaced by slab-sided, stub-nosed boxcar designs, their amorphous Lego-block hulls rising vertically above a keel. In their company she would have been an anachronism and so in slipping below, she paused as if to say goodbye, to us and to an era. She died as she lived — with class. We shall not see her like again.
Not until the ’72 ColRegs were adopted (1977) would the role and responsibilities relating to radar be clarified. At the time Andrea Doria appeared out of the fog, the Rules did little more than supply a precautionary note against placing too heavy a reliance on “scanty radar information.” Therefore this collision has to be seen through the lens of the times. The Rules governing then were those that went into effect Jan. 1, 1954 as Annex B of the 1948 SOLAS Convention and, as Farwell (Rules of the Nautical Road — U.S. Naval Institute) noted at the time: “contrary to some expectations, the 1948 Conference did not make special provisions for vessels equipped with radar.”
Eastbound Stockholm took the northern track against westbound traffic to save time and fuel with the master’s expressed belief that encountering shipping head-on was safer than crossing. (Studies done some years later by A.N. Cockcroft and published in the Journal of the Royal Institute of Navigation presented data showing that in restricted visibility, 75 percent of 494 collisions involved meeting situations and 20 percent crossing.
A major question to arise from the sinking was Doria‘s stability. The near-immediate list to about 20° virtually assured her eventual loss despite heroic crew efforts. The ability to withstand flooding of two adjacent watertight compartments was predicated upon any coexisting list being less than 20° — which in turn assumed periodic ballasting adequate to compensate for fuel depletion. (Ballasting to replace consumed fuel would necessitate pumping those saltwater contaminated tanks in New York before refueling — a time consuming and expensive procedure that was avoided if possible.) Should damage occur, and a list beyond 20° develop, those two flooded compartments would spill over to adjacent spaces. The outcome was sealed when the port intakes that would allow for counter-flooding were above water.
About the Author:
Following graduation from the U.S. Naval Academy, Jim Austin served aboard both a destroyer and cruiser with duties that included navigator, assistant CIC (combat information center) officer and air intercept controller. He subsequently worked on the submarine launched ballistic missile program for the General Electric Co.’s Ordnance Division. He holds a U.S. Coast Guard master’s license and writes frequently on ship collisions as seen through the twin lenses of the navigation rules and maritime law. He’s a retired physician living in Burlington, Vt.
