The Mystery of STENDEC

by Brian Dunning
November 9, 2010

It was a story borne out all too often in the annals of aviation disasters. An aircraft finds itself off-course and in the clouds with zero visibility, and worse, surrounded by mountain peaks that can't be seen. That's what happened on August 2, 1947, in the Andes Mountains of western Argentina. An Avro Lancastrian passenger plane of British South American Airways with 11 people on board struck a mountainside in zero visibility while descending toward what it thought was Chile. All aboard were presumed killed, with the crash not even being confirmed until more than fifty years later when its wreckage was finally found.

By themselves, the basic facts of what happened to British South American Flight CS-59 are tragic but not especially mysterious. The cause of the crash was a controlled flight into terrain. Nothing was broken, nothing was wrong, and it's unlikely the crew even saw the crash coming until the near-vertical cliff near the summit of Tupungato Mountain appeared through the mist. A second later, the Lancastrian was in a billion torn pieces, sliding down the face in an avalanche of its own making. Those pieces began to emerge from the bottom of the glacier, bit by bit, around the year 2000.

The mysterious part is what came over the radio just before the crash. In those days, long distance communication was by Morse code. The crew had been making regular hourly reports of its speed, position, and altitude, and all was well. At 5:41pm, four minutes before their anticipated landing time, the radio operator, Dennis Harmer, sent their ETA as 5:45pm and concluded it with the phrase STENDEC:

The Chilean Air Force ground controller asked him to repeat that, and Harmer sent it twice more, STENDEC, STENDEC.

And that was the last anyone ever heard of Flight CS-59. The mystery of the word STENDEC took its place among the great unsolved cases so beloved in the lore of urban legendry.

The Avro Lancastrian began its life as a British Lancaster bomber in World War II. Lancasters had four Rolls Royce Merlin engines, the front-line combat engine that powered the latest Spitfire and Mustang fighters. With some 5,000 horsepower on tap, Lancastrians were in demand for civilian applications after the war, such as the route over the Andes. The crew and passengers of Flight CS-59 had in fact just crossed the Atlantic from London on board an Avro York, a slightly different variant of the Lancaster; then changed planes to the Lancastrian for the final leg.

The JIAAC (the Argentinian Civil Aviation Accident Investigation Board) found the probable cause of the crash, but only after examining the wreckage upon its 2000 discovery. In CS-59's day, navigation was done by dead reckoning, making time/speed/distance calculations, as there were no radio beacons to help determine position. Flying east to west across South America, a plane must fight the jet stream head on. If the jet stream is faster than expected, dead reckoning calculations will lead you to believe that you've gone further than you actually have. Flight CS-59 had been in the air long enough to clear the Andes, and with no visual cues, began its descent into Chile, down into the clouds. Unfortunately, they descended right into a collision course with Tupungato. Once they entered its wind shadow, they fell victim to a phenomenon called a lee wave, part of which involves a downdraft over the face of the mountain, and at that point a crash became unavoidable. Without visibility, they almost certainly never saw it coming, and couldn't have done anything about it even if they had. These are precisely the same conditions that caused the 1972 crash of the Uruguayan Rugby team that inspired the book and movie Alive.

CS-59's crew were all experienced Royal Air Force pilots who had flown against the Germans in World War II. The pilot, Reginald Cook, had flown more than 90 combat missions. Cook's first officer Norman Hilton-Cook and second officer Donald Checklin each had over 2,000 flight hours. Dennis Harmer had served as a radio operator for three years during the war, and for over 600 hours for the airline. Among the crew, they had some 30 crossings of the Andes Mountains, though this was Cook's first time as captain.

Certainly such a crew would have been aware of the jet stream and lee waves, but none had ever negotiated the Andes before without an experienced captain. Although the cause of the crash was navigational error coupled with pilot error, the crew did about as well as anyone could have done, given the technology of the time, and the weather conditions nature thrust them into.

But the 2000 finding of the true cause of the crash has not stemmed the speculation about what significance STENDEC had. Theories even extended to UFOs, with some people proposing the cryptic message must have been some kind of warning about the plane's abduction by a giant alien craft. Even aviation medical experts have gotten into the fray, suggesting that the crew may have been suffering from hypoxia. Lancastrians were not pressurized but the crew did wear oxygen masks, and perhaps Harmer was not wearing his or it had malfunctioned. But if this was the reason for his garbled transmission, it seems unlikely that he could have repeated it verbatim, twice, when asked by the Chilean controller.

All kinds of people have made all kinds of guesses about what STENDEC might have meant. None of them are very convincing. Most interesting is that the letters are an anagram of DESCENT, and Flight CS-59 was certainly on its descent. Did Harmer simply miskey the letters of DESCENT? It seems to be quite a stretch that such an experienced operator would have made the exact same extreme misspelling three times in a row. Moreover, it would have been a totally random place in the transmission to throw in this word. Nobody ever did that.

It's also been proposed that Harmer miskeyed the Aircraft's name, which was Star Dust (all of British South American Airways' aircraft were given two-word names beginning with Star). This possible misspelling has been a popular theory, so much so that most stories you'll find about Flight CS-59 are titled The Mystery of the Star Dust or some such thing. However, in reality, nobody at the time ever referred to the aircraft or the flight by the unofficial name painted on it by the airline. Probably the crew was scarcely even aware of the plane's name. In radio transmissions of the day, planes were only ever addressed by registration number, in this case G-AGWH. And, once again, it would have been completely random and uncalled for to tack the plane's name onto the end of a routine transmission.

Most other theories are that STENDEC was an acronym of the first letters of words in some sentence, and various nominations have been proposed. None of them make sense, because no radio operator would ever communicate in such a way, and certainly not repeat it twice more when told that it was not understood.

One fairly bad match is "Santiago tower, now descending entering cloud", which is true but no sane radio operator should expect such an acronym to be understood. Another is "Star Dust tank empty no diesel expecting crash", which has five problems: Harmer would not have identified the aircraft by its unofficial name; Star Dust was two words, not one; Merlin engines do not burn diesel; the plane was not low on fuel; and Harmer was experienced enough to know there would be no reasonable expectation of anyone comprehending such an acronym.

Morse code experts have searched for a solution within the subtleties of hand-keyed code. There is an attention-getting signal consisting of three dots followed by a dash:

...and these are the same as STENDEC's first two letters, except that they have a space. Listen for the short space between the dots and the dash:

Morse code also uses something called prosigns, which are two letters run together without a space. The prosign AR, which means "End of message", sounds like this:

...and that's the same as STENDEC's final two letters, except that, again, they have a short space after the first dot:

So, theoretically, STENDEC could mean ATTENTION - END - END OF MESSAGE. Now, this sounds fairly reasonable. And it could be, except that it wouldn't make much sense for Harmer to send this. STENDEC followed his transmission of their estimated arrival time, so it would be rather strange for him to insert the ATTENTION signal in the middle of a message. Similarly, it would be redundant to say END and then transmit END OF MESSAGE. None of Harmer's other transmissions from the flight did this, and no radio operators of the day ever did, so far as I could find. I think we can call this the best effort from the Morse code experts, and it shows that there is probably not a good answer to STENDEC to be found within the imperfections of hand keying.

Addendum: For the most plausible alternate explanation of STENDEC I've seen, see John Scherer's 2010 article in the North Texas Skeptics newsletter, which points out that the keying of STENDEC is very similar to VALP, representing a possible alternate landing site at Valparaiso. —BD

Another addendum: A different explanation was proposed by listener Dean, and detailed in this feedback episode. —BD

Is it really necessary to assign some profound significance to STENDEC? What about all the other meaningless, garbled, or unclear radio transmissions that happen every day, that nobody remembers because they're not followed by a fatal crash? I've flown a lot, and I think every time I go up I hear someone say something I can't make out. I've no reason to suspect it was any different in the days of Morse code. Who knows what Harmer meant to say? Most likely, he was about to continue with what would have been a meaningful transmission when it was cut short by the crash. Most of these guys were in their twenties and Harmer could have even just been horsing around. The bottom line is that the cause of the crash had nothing to do with anything Harmer sent or didn't send on the radio. It would have crashed anyway, and chances are that if Harmer had lived five minutes longer he and the Chilean air traffic controller would have connected a little better, and the miscommunication never would have been remembered.

I see no reason to link the miskeyed, misread, or made-up transmission with the crash. The only thing connecting them is that they happened within a few minutes of each other, but the crash is simply not missing any explanations requiring an interpretation of STENDEC. Consider it an unsolved little tidbit from history, but one whose significance is almost certainly limited to the interest it holds as a puzzle.

By Brian Dunning
Follow @BrianDunning