-
The accident plane, ZK-NZP coming in to land at Heathrow Airport, 08 April 1976 - Source: Werner Fischdick -
Related
Chief Inspector of Air Accidents Ron Chippindale concluded his
1980 accident report into the Erebus tragedy by stating the
accident's probable cause was, "the decision of the Captain to
continue the flight at low level toward an area of poor surface and
horizon definition when the crew was not certain of their
position&" (Paragraph 3.37) He further asserted that the flight
engineers had expressed "apprehension" and "dissatisfaction" at the
flight's "continued descent towards a cloud-covered area."
(Paragraphs 2.25, 3.24). The basis for these claims was the content
of the cockpit voice recorder ("CVR") transcript appended to the
report.
Investigating the Erebus cockpit voice recorder: a critique
The following is a critique of the published official cockpit voice
recorder transcript from ZK-NZP, the DC10 aircraft involved in the
accident at Mt Erebus, Antarctica, on the 28th of November, 1979,
in which all 257 persons on board lost their lives.The critique
provides an interesting study in what can happen when established
and agreed upon investigative procedures and evaluation techniques
are not subsequently followed.
In the following article, Captain Gary Parata (Chairman of
NZALPA's Accident, Incident and Safety Group and a flight recorder
specialist with the IFALPA Accident Analysis and Prevention
Committee) considers flaws in the procedure used for the production
of the published transcript, and why - as a result of those
procedural flaws - the Chief Inspector's theory and conclusions
were misleading.
Background information
A cockpit voice recorder or CVR is a device that, as its name
implies, records ambient voice (and other sounds) present on the
flight deck of aircraft.
It also records all passenger address announcements and radio
communications between the aircraft and ground stations. Unlike
modern solid-state units, the CVR fitted to ZK-NZP consisted of an
earlier generation magnetic tape, contained within an
impact-resistant module. It incorporated an 'endless-loop'
mechanism, with conversations recorded more than about 30 minutes
earlier being continuously recorded over by more recent data.
The CVR unit was found in moderately good condition, with the
impact-resistant module not seriously damaged. The unit was sent to
Sundstrand, the manufacturers of the unit in Seattle, Washington,
in the USA, after the accident, where it was carefully disassembled
and a working copy of the tape produced.
Much of the intra-cockpit communications were recorded only via a
low-fidelity (but high sensitivity) cockpit area microphone mounted
in the overhead panel between the two pilots. While both pilots'
voices were, for the most part, loud and clear, conversations
taking place behind them were muffled, indistinct, and sometimes
simply not intelligible. The sensitivity of the system was so high
that voices from the forward galley area aft of the flight deck
were also recorded. It would be accurate to say that this CVR
system was not designed to accommodate a scenario where several
persons positioned remotely from the cockpit area microphone might
be speaking and be recorded. Despite these issues, it was still
possible to obtain a good quality transcript from most CVR,
provided certain protocols were employed for doing so.
The process
Building an accurate and authentic transcript from a CVR, and then
correctly interpreting the results, can be a painstakingly slow and
tedious task. Specific protocols are followed in order to prevent
dubious or unclear passages from being transcribed as definitive
phrases. Dedicated audio equipment able to filter out extraneous
noise and enhance other sounds must be employed in order to give
reliable results where recordings give poorer-quality data.
Once transcription is complete words and phrases must be read, first in isolation, and then from within a broader context, in order to either support or contradict any initial interpretation. A common trap is to ignore this broader context and assume that the first interpretation, made without contextual reference, is the correct and authentic representation. It is almost inevitable that this approach will lead to an erroneous conclusion. In fact, it is normal when constructing a transcript to state at the beginning of the transcript itself: Never draw conclusions, nor ascribe meaning to a phrase, using the CVR as the sole source of information.
To illustrate this, imagine recording a meeting with five people conducting two simultaneous conversations. This scenario is not unlike the environmental situation on the flight deck of a large 1970s-era aircraft with flight engineers and perhaps navigators present as well as the pilots. Then imagine asking five untrained people working independently to play back the tape on their own personal equipment and transcribe their results. They will tend to write what they think they hear, or worse, what they want to hear, rather than what is actually said. Moreover it is likely there will be five widely differing results, any or all of which may be misleading.
Without the services of an audio expert it is quite easy, for
instance, to blend two simultaneous conversations into one and thus
reach a completely erroneous result.
Despite the poor quality, the lay person could be persuaded by
eloquent argument that any (or indeed all) of these "transcripts"
are correct, even if they contradict one another. This is a good
example of the Rashomon Effect 2, in which one person's truth
conflicts with another person's equally convincing truth, and is a
well-known phenomenon in psychology.
For this reason, properly trained CVR analysts will do only two things when constructing a transcript: they will attempt to determine exactly what words were spoken, and will attempt to attribute those words to a particular person. It is absolutely critical that transcribers do not attempt to determine why something was said, nor attempt to interpret or infer any meaning, during the transcription process. That will come later, when the transcribed words are analysed and interpreted alongside any corroborating information.
Good rules and protocols for CVR transcription and interpretation are essential for forming strong defences against flawed results. They facilitate the production of a true and authentic record, and promote an accurate understanding of the events in question. However even with these precautions in place, highly experienced air accident investigators are still not immune from hearing the things they want - or expect - to hear on a CVR. Nor are they immune from ascribing meaning to a phrase based only on supposition, or perhaps subtle pressures from vested interests.
They must therefore always remain vigilant, and flight recorder
specialists are trained to
understand this. In order to assist investigators further, it is
axiomatic that transcribers and interpreters must be available that
are in current flying practice with the same airline, and on the
same type of aircraft, as was involved in the accident.
Moreover, they must not work alone or without the assistance of trained audio specialists. If there is significant disagreement among transcribers, or if none are able to determine what was said in relation to a particular passage, that particular passage must be deemed to be unintelligible, or not sufficiently intelligible to give a reliable result. In addition, anyone working on an investigation or a transcription should remain free of preconceived opinions or ideas as to possible causes of the accident. This will guard against the brain being fooled into hearing something that supports a pre-conceived theory, rather than what might actually have been said.
The initial CVR Group assembled by Mr R Chippindale travelled to Washington and underwent specialised training, adhered strictly to that training, and produced a single, handwritten transcript. For their efforts and professionalism they were roundly praised by NTSB and FBI experts.
The transcript actually published in the formal accident report was significantly different to the version produced by the CVR Group. Chippindale made 55 changes to the transcript without consultation with the CVR Group and in direct contravention of accepted protocols. Chippindale's actions were inexplicable - 'at best, an extraordinarily non-standard performance; at worst, a highly improper and prejudicial act.'
The complete failure of the accident-investigation system to identify and prevent unprofessional conduct like this was what Justice Mahon later described as "culpability of the organisational system" (Mahon, 1981).
Chippindale's involvement
In the initial stages of this investigation the
Investigator-in-Charge ("IIC"), who was New Zealand's Chief
Inspector of Air Accidents, Mr R Chippindale, properly convened a
team of appropriately qualified persons under the supervision of an
Inspector of Air Accidents from his office, to sit on a CVR
"Group." This was one of a number of panels that was formed to
examine specific areas of the accident. It was the job of this
group to build the CVR transcript.
As well as the Inspector, the team members consisted of the airline's Fleet Manager DC10, Captain Barney Wyatt, the Chief Flight Engineer DC10, Don Olliff, and a representative from the New Zealand Air Line Pilots' Association, DC10 Captain Arthur Cooper. These three were employed by the airline and well acquainted with the accident flight crew. To carry out the transcription this team took the working copy of the CVR tape to the American National Transportation Safety Board audio laboratory in Washington DC. Prior to commencing their task they received specialised training in the audio transcription protocols from NTSB flight recorder specialists and FBI audio technicians.
After some five days, during which their mentors carefully
monitored progress, they produced a single hand-written transcript
3. Cooper recalls the Americans were impressed with the patience
and professionalism shown by the New Zealanders. The supervising
Inspector then briefed the others on what their tasks would be upon
their return to New Zealand; the creation of a single typewritten
version, certification of this copy as a true and complete
rendering of the hand written original, reproduction and
distribution of the certified copy to the investigation
participants, and the provision of expert on-going advice,
analysis, interpretation, and editing.
Their transcript was to be represented as the definitive article
that would form an integral part of the investigation.
The CVR team returned to New Zealand with the plaudits of the Americans still ringing in their ears for a job well done. It would be fair to say that they expected their work to be accepted as it stood.
On the team's return, the CVR Group reported back to the IIC on the work done and result achieved, fully anticipating to continue supervision of the single handwritten transcript and its workup into certified typewritten form. But sometime thereafter, in the first of what would turn out to be a series of departures from normal practice, the IIC took complete and sole control of the transcript and audiotape. The CVR Group did not realise this for some time.
The IIC subsequently took the audiotape and transcript to his home, and in another departure from internationally accepted practice (that to an investigator can only be described as inexplicable), conducted a private CVR meeting at which it appears only one other person was invited - the airline's management representative, Ian Gemmell (Macfarlane 1991, 308). What transpired at this meeting, which under oath both the IIC and Gemmell confirmed took place 4, is not known.
There were no valid reasons to hold the meeting. The IIC had a group of specialists already convened to answer any questions he, or the other members of the investigation as a whole, had pertaining to the CVR. Any issues the IIC had with the Group's work should have been referred back to them, not to Gemmell.
In any event, while Gemmell possibly could have brought to the
IIC some additional knowledge
about CVR, as an airline representative he was biased toward a
"pilot error" position. It is impossible to say whether Gemmell
used his presence to influence what the IIC was hearing on the
tape, but the opportunity for this to have happened at that point
in time was significant. Such a possibility should have been
prevented by the IIC to avoid an appearance of personal bias and
improper conduct.
Sometime after this meeting, the IIC took the tape to an alternative audio laboratory at Farnborough, England (Mahon 1981, 37). He did not follow established safe-guards and protocols in doing so, listening to the tape only in the company of Farnborough officials. None of those officials held DC10 qualifications, were personally acquainted with the crew or were familiar with the airline's DC10 operation.
Throughout this time the original CVR Group believed that the normal process, as briefed by their supervisor, was under way and that in time the typewritten transcript of their work in Washington would be ready for their careful checking and certification before it would be utilised. They were not aware that their involvement in the investigation had in fact ceased. Cooper, and most likely the other members of the CVR Group, were in fact kept in the dark until the publication of the final report.
Cooper was astounded to note a typewritten transcript at Annex 'C' of the final report that he had never seen, let alone certified. Cooper reviewed the CVR Group's transcript 5 and compared it to the published version. He became concerned that it appeared to contain additions, deletions, and editorial work that was clearly not from the CVR Group. Eventually it would be determined that there were 55 differences between the original work and this alternative transcript.
Clearly then, the IIC had had cause to interrupt the planned
transcription process.
Firstly, he must have thought the CVR Group's result was wrong,
incomplete, or needing his assistance in some way. What assistance
he could have given to his CVR specialists is not clear. He must
have had knowledge that was not known to, or acted on by, the CVR
Group. For example, he may have had information that indicated a
better result could have been obtained at Farnborough, as opposed
to Washington. This might be understandable if he had a concern
about the number of passages deemed unintelligible by the CVR Group
in Washington. He is not on any record as having expressed this
concern.
In any event, he should have informed the Group of any issues he may have had and sent them to Farnborough to re-evaluate. He did not do so. He travelled to Farnborough himself, did not involve any of his CVR specialists, and produced an alternative transcript that, in places, bore no resemblance to the CVR Group's work. Some of the differences were highly contentious, and these are examined below.
The IIC had substituted the original transcript constructed by the CVR Group, with all of its checks and balances, with his (and possibly Gemmell's) own. At best, this was an extraordinarily non-standard performance. At worst it was a highly improper and prejudicial act. Either way it should not have been allowed to happen. That it did speaks volumes for the weakness of the process, and the lack of any management oversight of the Office of Air
Accidents Investigation at that time
The CVR Group supervisor's opinion of what transpired is not
recorded. On 31 May 1980, the IIC's report was published, and
shortly thereafter released to the public.
The changes Mr Chippindale made suggested that he had predetermined the cause of the accident and manipulated the CVR evidence to support his theory that pilot error caused the accident. This is evidenced by the fact that Mr Chippindale expressed an expectation of a verdict of pilot culpability to the families as early as 14 December 1979, a mere two weeks after the accident.
Justice Mahon, in his report for the Royal Commission of Inquiry, discounted the validity of Mr Chippindale's CVR edits, however he stated that: "I am satisfied that there was no deliberate editing of the transcript so as to conform with the Chief Inspector's opinion&. [But] the Chief Inspector had a natural inclination to ascribe to remarks of doubtful meaning an interpretation which favoured his own theory because, believing as he did in the validity of that theory, he also believed that members of the flight crew must from time to time have expressed apprehensions." (Paragraph 121)
Changes made by the Investigator-in-Charge
The most controversial of the departures from the original
transcript was the addition of this infamous phrase, which was
dealt with at length in the Mahon Report:
Time 0047:55 bit thick here, eh Bert?
This conjures up images of poor visibility and of an aircraft being operated incompetently and about to stray into danger. The IIC claimed the words were definitely spoken to a flight engineer by an unknown person. The problem is that no other person is known to have heard it. Certainly, none of those words came from the CVR specialists. The phrase simply was not sufficiently intelligible for them to transcribe in accordance with the training they had received (Mahon 1981, 37-38).
Several factors can be taken into account to assess the validity of the transcribed phrase:
1. The CVR Group, accompanied by a number of experts, could not
agree on any words being clear enough in that passage to issue a
definitive transcription. When Mahon J went to the Washington
laboratory and listened to the tape in the company of specialists
with credentials similar to the CVR Group, they all agreed that the
phrase was far too garbled to be given any reliable transcription.
Specifically, Mahon J discovered that a word sounding like "here"
was not heard by him or any of the other specialists present; the
speech sounds were vaguely similar to "bit thick eh Bert" (Mahon
1981, 38).
2. The IIC claimed that he and a Farnborough specialist heard the
words. When Mahon J himself went to Farnborough he discovered that
the "specialist" was a "filtering expert", a flight recorder
technician working in the audio laboratory whose job it was to
operate the lab's equipment (Mahon 1981, 38-39). There were several
qualified investigators present with Mahon J and the technician
when he listened to the recording there, including the Chief
Inspector of Air Accidents for the United Kingdom. The specialist
investigators at Farnborough declined to transcribe the phrase as
it appeared in the published form, and this time the technician did
not give an opinion at all.
In later evidence 6 the IIC claimed that Mahon J heard the word "here" as "there" and he (the IIC) "&would now tend to agree." This difference in words changes the whole meaning of the phrase, and negates any implication that the aircraft was flying in cloud. Mahon J wrote that neither he, nor Counsel assisting him, nor the Farnborough officials had heard any such word (i.e. "here") at all.
In a newspaper article published in 2004 the then-retired Chief Inspector was challenged, and admitted he may have got the name "Bert" wrong. "But the 'bit thick here' comment was there," he said (Loughlin 2004).
3. It was certainly possible that the IIC made an genuine error
in reaching this result. In that case, there was no mechanism in
the system that could have identified and rectified the error. This
was because he had taken sole control of the CVR and no-one was
able to monitor him.
This illustrates why no single person ought to be permitted to
determine the outcome of any CVR investigation. Nor should they
decide what should and should not be included in any CVR
transcription, including during editing.
4. The IIC said that a question was being asked of the flight engineer, referred to as "Bert." He knew that the flight engineer's name was not "Bert." In fact, there was no flight crew member named "Bert".
5. The word "thick" strongly implies the aircraft was flying in or near cloud. However, passengers' photographs from cameras recovered from the crash site did not support this; a number of photographs showed sunlight streaming into the cabin right up until impact (Vette and Macdonald 1983).
These failures mean that the IIC acted beyond proper and appropriate boundaries when he submitted these words. They should not have been transcribed.
Any one of the failures identified during the transcription process would be enough to cause the CVR Group to correctly render the phrase as unintelligible. Alternatively, during the interpretation phase of the CVR study it would have been adjudged as: not sufficiently intelligible to be given any reliable meaning or interpretation.
In a press release made at the time the IIC claimed that: "Just because there was no crew member named Bert that doesn't prove that the words weren't spoken. I knew they were from the long hours spent listening to the tape" (Macfarlane 1991, 329). This statement reveals another flaw in the process with which the IIC transcribed the CVR. Audio engineers and flight recorder analysts are very careful not to spend long periods, especially alone and unmonitored, trying to understand words. As a trainee flight recorder specialist learns early on in their career:, after a while the brain will begin to hear things that aren't there.
"Bit thick eh Bert," was far more likely to be: "This Is Cape Bird"
There was strong evidence to suggest that the words spoken were probably: "This is Cape Bird." Based on the route shown during the Route Clearance Unit (RCU) briefing the crew may have believed that it was Cape Bird visible on the left hand side of the aircraft at the time these exchanges took place, with the next landmark to come into view on the left expected to be Cape Royds. In fact it would not be Cape Royds, but rather Cape Tennyson, which was probably subtending the same angle and thus a similar visual aspect as might have been expected from Cape Royds (Vette and MacDonald 1983: 134).
How the crew mistook Cape Tennyson for Cape Royds (after Vette
and Macdonald 1983).
The words "I reckon Bird's through here" were clearly spoken
seconds afterward, as confirmed by the CVR Group. Contextually, the
phrase would not be out of place. However mere suggestion that the
words spoken were "This is Cape Bird" fails the test required to
insert them as a definitive passage 7 into the transcript.
In other words, 'probably' isn't good enough; the only acceptable test is 'definitely'. And the only acceptable method by which 'definitely' can be achieved is by way of a vote, of a group of specialists that themselves meet certain criteria. Clearly then, the treatment of the CVR failed to meet these criteria. The IIC failed to show that his version of the phrase was spoken. Thus, the original conclusion reached by the specialist CVR Group that the words are unintelligible, could be the only correct result.
Conclusions
The original CVR Group correctly constructed a CVR transcript (the
"Washington Transcript").
The IIC produced a second, non-authenticated transcript (the
"Farnborough Transcript") containing no less than 55 departures
from the "Washington Transcript" produced by the CVR Group.
The "Farnborough Transcript" was then substituted for the correct version in a manner quite unique to the annals of air safety investigation.
The CVR Group was not apprised of the substitute transcript prior to its appearance in the published accident report.
The "Franborough transcript" was able to be published and represented as an authenticated record.
The published transcript contains passages that do not meet any criteria for inclusion in such a record, including the infamous additions creating false images that the aircraft was flying in poor visibility and that its crew was lost.
The published transcript omits certain passages authenticated by the CVR Group, that show the flight crew were, as far as they knew, operating their aircraft properly.
The additional work carried out by the IIC was not subject to any form of independent quality control or checking and therefore ought not to have been relied on in any way.
The behaviour of the IIC throughout is consistent with his focus on one theory, that the crew were unsure of their position and flying in poor weather conditions. As a result the published transcript reflects that personal belief, rather than providing and unbiased conclusion.
Captain Gary Parata - About the author
Captain Gary Parata is an airline pilot with over 12,000 hours'
experience. He is an accredited air accident investigator with the
International Federation of Air Line Pilots' Associations (IFALPA)
and Chairman of the Accident, Incident, and Safety Group of the New
Zealand Air Line Pilots' Association. He is also a flight recorder
specialist with, and Vice-Chairman of, the IFALPA Accident Analysis
and Prevention Committee. In November 2005 he was appointed as
IFALPA's representative to the ICAO Accident Investigation Group
and in May 2007 as its representative to the ICAO Flight Recorder
Panel.
Acknowledgments
The author would like to acknowledge the following in the
production of this article: retired DC10 pilot and NZALPA Scroll of
Merit awardee Capt. Arthur Cooper, whose first-hand account of the
CVR Group's activities was invaluable; Stuart Macfarlane, retired
lecturer in law at the University of Auckland and editor of The
Erebus Papers, who provided much needed and on-going clarification
with a number of legal and factual issues; Craig Oliver, former
technical officer of the NZ Air Line Pilots' Association, who
suggested some editorial amendments; Anne Cassin (a pilot herself)
who lost her husband that day, and whose recall of events in 1979
is still clear; NZALPA Life Member Capt. A. Gordon Vette, ONZM
HonDEng (Glasgow) JP, whose selfless sacrifices and dedication to
the profession of airline pilot shine like a beacon, exhorting us
all to ever greater things; Capt. Paul McCarthy, whose analyses of
the many drafts was necessarily brutal; and finally to those that
were lost, and their families and friends, because they are the
real reason that we continue to study this accident. It is for
them, and indeed for all of us who fly, that we must ensure that
all the lessons learned are not lost, too.
References
Loughlin, D., (2004) Transcript Casts Doubt on Erebus Report.
Wellington: reported on 4 December 2004 in the "Dominion
Post"
Macfarlane, S., (1991) The Erebus Papers. Auckland: Avon
Press
Mahon, Hon P.T., (1981) Report of the Royal Commission to inquire
into the Crash on Mt Erebus, Antarctica, of a DC-10 aircraft,
operated by Air New Zealand Limited. Wellington: Government
Printer
Office of Air Accidents Investigation, (1980) Report No 79-139.
Wellington: Government Printer
Vette, A.G., and Macdonald, J., (1983) Impact Erebus. Auckland:
Hodder and Stoughton
For more information visit the
NZALPA Erebus website
