Concerning Saturday's front-page story in the NYT about what the Teterboro air traffic controllers were doing just before the airplane-helicopter collision over the Hudson:

Obviously this doesn't look good for the controllers (that one of them was on a "non-business-related phone call" just before the crash), and the National Transportation Safety Board will eventually pronounce on how much, if anything, that had to do with the crash.* The NTSB's special update on what it has learned so far about the controllers' behavior and other factors is here.

There's one main reason to think that none of the controllers, including the one in the Teterboro airport tower who was on the phone, should principally be "blamed" for the crash. The reason is that by definition controllers are never principally responsible for "traffic separation" when planes are operating under "Visual Flight Rules," or VFR. The pilots themselves are responsible, like the drivers of cars.

When the weather is clear and pilots are operating under VFR, they are free (within limits) to choose their own course and altitude; but they -- not the controllers -- bear legal and practical responsibility for staying clear of terrain and not running into anything else in the sky. Everyone involved in the system understands this. The big divide in aircraft operations is between VFR and IFR, "Instrument Flight Rules." Under IFR, the pilots have to go where the controllers say -- but the controllers bear legal responsibility for keeping one plane away from others. Virtually all airline flights operate under IFR, so non-pilot public assumes that controllers are supervising flights of every kind. They're not.

Also, based just on the facts now released, there's something to be explained about the airplane pilot's actions.  Soon after the plane had taken off, the Teterboro tower controller told the airplane pilot to switch to a Newark "departure" controller on another frequency. This is purely routine and is something you expect once airborne from an airport with a control tower. ("Airplane XXX, contact departure on [ XXX frequency]".)** Usually you know ahead of time what frequency you'll be switched to, and you have it pre-loaded into your radio. When instructed, you activate that new frequency by pushing one toggle switch.

In this case, the pilot acknowledged the "contact departure" request but then never spoke to the new controller. "Never" covers the 54 seconds between the request to switch frequency and the actual crash. That's a pretty long time not to "check in" with the next controller. Usually you enter the new frequency (a few seconds); listen a few more seconds for a chance to talk; and then announce yourself to the new controller. In extremely busy air-traffic areas, like New York most of the time, you may have to wait quite a while for a break in transmissions so you can check in. Was the pilot waiting all that time? The tapes will show whether he had a chance.

Now we come to the area of murk and "responsibility" in other than a strictly legal sense, which the NTSB will try to sort out. The NTSB announcement says that the second controller, in Newark, was eager to reach the pilot to warn him about the helicopter and suggest that he turn to avoid its path. Obviously that warning never get across. Was it just because the frequency was too busy? That would seem odd: when a controller really wants to reach a particular plane, he can tell other pilots to be quiet and put out a call to the plane he needs to reach. It's not unusual to hear such instructions. ("Piper XXX, if on frequency, acknowledge; all other traffic stand by....") Did the controller ever put out such a call? The NTSB doesn't mention it, but says that the Newark controller telephoned the one in Teterboro to mention the problem. Of course that Teterboro controller could no longer reach the pilot, whom he had instructed to switch away from his frequency.

The lore of aviation disasters, often discussed here, is that they very often involve an "accident chain" that could theoretically have been broken at any link. If the Teterboro controller had not been on the phone, maybe he would have seen the same impending problem that the Newark controller did -- but maybe not, because his radar scope may not have covered the same area. If the pilot had been able to check in quickly with the Newark controller, maybe he would have gotten the warning and turned. If the Newark controller had tried to reach the pilot, maybe that would have paid off.  If the pilot and passengers had been looking in a different part of the sky, maybe they would have seen the helicopter in time. We'll know more about this eventually, although the whole tragedy may never be fully explained.

Main point for the moment: it would be natural for non-flying readers to hear about the controllers and conclude: Obvious negligence! They should have been at the scope keeping those planes apart! That's been the implication of some recent coverage of the crash. It is indeed possible, based on what's known now, that controllers might, through extra vigilance, have averted this disaster. The one in Newark apparently tried. But this is different from a situation in which, say, a controller neglects his duty to keep airliners safely separated and allows them to collide. Here what we know so far is that controllers may have missed a chance to go beyond normal duty and save the pilots from error. More when the NTSB speaks.

UPDATE: According to this AP story, as of Monday night the NTSB revised a previous claim that the Teterboro controller (the one with the phone call) could have seen the impending collision. The new info suggests that the helicopter did not show up on that controller's screen until immediately before the crash. Main reminder: it will take a while to sort out what really happened.

*Side note: for all the care and thoroughness of the NTSB, its final reports can be weirdly tautological. If a plane has crashed on takeoff, the finding of probable cause may talk about the "pilot's failure to maintain proper terrain clearance." When it eventually reports on this Hudson crash, the conclusions will probably include both pilots' "failure to maintain proper separation from other traffic." Still, it does careful, exhaustive work, and its reports end up containing as many crucial facts as can be found.

**Why is a pilot talking with controllers at all, if he is flying VFR? There's a very long answer, but the short version is: at airports with control towers and in certain categories of airspace, a pilot must be in radio contact with air traffic control and obey their instructions -- even if operating VFR. So even if this pilot was planning a VFR trip out of New York, which meant that he would choose his own course after he got away from the city and would be responsible for seeing and avoiding other planes,  he needed to talk with controllers in these early stages of the flight.

We want to hear what you think about this article. Submit a letter to the editor or write to