An analysis of the recording of Armstrong’s statement on stepping on the Moon in July 1969.
Neil Armstrong’s statement made on setting the first human footprint on the Moon in July 1969 has been widely praised as one of the great moments of spontaneous poetry in human history. Whilst the meaning of his declaration “that’s one small step for [a] man, one giant leap for mankind” is not in doubt, much has been written about the actual words recorded back here on Earth and the apparent absence of the article “a” before “man”. In an analysis of a high quality uncompressed first generation recording of the mission audio recently released by NASA and other archive recordings of Armstrong speaking before, during and after the mission, we present cutting edge research evidence from Linguistics in support of the speech being spontaneous rather than a recitation of something pre-written and we provide phonetic evidence that the “a” was intended, even if it is absent in the recording.
At 02:56 UTC on July 21st, 1969, six and a half hours after landing the first manned spacecraft in history on the Moon’s surface with Buzz Aldrin, 39 year old Apollo 11 Commander Neil Armstrong from Wapakoneta, Ohio, wriggled out of the spacecraft hatch on his belly and slid down the slight incline of the Lunar Module (LM) porch towards the top of the ladder. With his left hand he pulled on a D-ring hanging near by to unfurl the Modular Equipment Stowage Assembly (MESA) folded against Eagle’s side near the base of the ladder and upon which was fixed a single black and white TV camera.
Whilst picture engineers across the world rushed to improve the image Armstrong reached the bottom of the ladder and dropped down the final few feet and on to the footpad at the end of the LM’s leg. His historic broadcast began with a description of his limited, but unique view of the Moon’s surface from near the base of the spacecraft. The picture improved a little as this first explorer from Earth relayed his first impressions of the properties of the lunar dust. Armstrong then collected his thoughts and announced he was going to step off the LM. Adjusting his balance he shifted his left foot onto the Moon’s surface, taking the next 10 seconds to contemplate the moment privately before declaring; “That’s one small step for a man, one giant leap for mankind”.
His words were picked up by a microphone inside his clear polycarbonate pressure helmet, mounted near his lips and held in place by his black and white “Snoopy cap”. From here the words were carried by VHF radio signal through a small aerial mounted on the top of his personal life support system (PLSS) back pack and up to an aerial on the roof of the Lunar Module. From the LM’s main communications system it was relayed back to Earth through a roof mounted S-band dish at 2.2 GHz.
Back on Earth the S-band signal was picked up by one of NASA’s 26 meter diameter deep space network dishes, at Honeysuckle Creek in Australia, which had a line of sight to Tranquillity Base. Two additional 64 meter diameter dishes at Goldstone in California and Parkes in S.E. Australia had also been brought in to help receive the TV pictures, Lunar Module telemetry and voice feeds during the Apollo 11 Moonwalk. Signals from these ground stations were relayed to Houston through a network of ground cables, sea cables, microwave links and Intelsat geostationary satellites collectively known as the NASA Communication Network (NASCOM).
The NASCOM was managed by the Goddard Space Flight Center in Maryland which received the voice, telemetry and tracking data before it was sent on to Houston. The Apollo TV signals went direct to Houston for onward transmission to the broadcasters.
An estimated six hundred million people listened and watched around the world, marvelling at the moment as commentators on dozens of TV channels repeated Armstrong’s spontaneous line as they had heard it; “that’s one small step for man, one giant leap for mankind.”
Sure enough when Armstrong’s actual statement was replayed back on Earth in the hours which follow it appeared to be missing the crucial word “a”, apparently changing the entire meaning of the carefully chosen words. In the months and years which followed numerous press articles and reports debated the controversy.
Armstrong has always insisted he spoke the “a” and NASA have suggested that perhaps it was somehow lost en route to Earth in the downlink or obscured by static interference.
Research conducted since 1969
In Andrew Chaikin’s 1994 book, “A Man on the Moon”, Armstrong is quoted as admitting that he had of course intended to say “one small step for a man” and believed that he had done so. In James Hansen’s 2006 biography of Armstrong, “First Man”, Armstrong is quoted as acknowledging that maybe he did get the sentence wrong; conceding “it doesn’t sound like there was time for the word to be there. On the other hand”, he continued, “I didn’t intentionally make an inane statement, and …certainly the ‘a’ was intended, because that’s the only way the statement makes any sense.”
Then in September 2006 Australian journalist and entrepreneur Peter Shann Ford announced that he had found space for the missing “a” in the waveform of Armstrong’s transmission . Ford concluded that it had originally been spoken too quickly to be picked up and transmitted back to Earth, but this new analysis had still been able to detect the signature of the missing word.
According to Ford the “a” had lasted just 35 thousandths of a second, 10 times too quickly to be heard. But the shape of the sound wave, Ford felt, was consistent with the sound made by the tongue, mouth and lips as the speech transitioned from the final phoneme /r/ in “for” to the word “a” and on to the opening phoneme /m/ in “man”. Ford concluded that such a pattern would not be present had Armstrong moved straight from the “r” in “for” to the “m” in “man”, as he did later, with the phrase “for mankind” which lacks the signature Shann detected earlier for the “a”.
Subsequent criticism [2,3,4] and more detailed analysis of the audio’s waveform and spectrogram by other researchers on the Language Log website [5, 6],and from audio engineer Garth Wiebe  found no such trace of an “a” or any other vowel sound within the crackle in the gap between “for” and “man” and concluded that the audio moves evenly from the “r” sound to the “m” sound in the same way as it does for the words “for mankind”.
Although largely unreported since the widespread coverage of Shann Ford’s original 2006 research, the truth is that there seems to be no trace of the “a” in the recordings.
With this in mind, and as suggested in some of the responses to Shann Ford’s work noted above, we have set out to offer some linguistic and phonetic explanations for why the “a” appears to be missing, drawing on other examples of Armstrong’s speech not only from the mission itself, but also from archive recordings made before and after the flight of Apollo 11.
In addition we have also examined outside influences on Armstrong’s use of language ranging from the dialect region of North America he grew up in, to the speech patterns of his parents recorded in interviews from the 1960s. Finally, we present a phonetic analysis of Armstrong’s historic statement and draw some new conclusions from it.
The archive we have used
For this study we have utilised archive recordings of Armstrong’s voice from the Apollo 11 pre-flight press conference, a selection of mission downlink audio samples from the editor of NASA’s Apollo Flight Journal, and a post-mission interview done by astronomer Patrick Moore for a 1970 episode of the BBC’s long running The Sky at Night series. Our mission audio is the best released by NASA to date – taken from the original magnetic tape recordings made at Houston which have recently been re-digitised to make 48 kHz, 24-bit-precision uncompressed digital audio wav files, compared to the earlier, heavily compressed 24kbps stream rate of the mpeg digital files which previous researchers have used.
The samples of Armstrong’s parents’ voices we have used were recorded by CBS on a 1962 episode of I’ve got a Secret – which can be found today on Youtube, and from an American network news interview recorded during the night of the first Moonwalk outside their home in Wapakoneta, Ohio.
Armstrong’s speech performance
After carefully listening to Armstrong’s speech during the pre-flight press conference, throughout the mission itself, and in the post flight BBC Television interview we have concluded that Armstrong generally suffers no speech production problems of any kind, including any tendency to drop the word “a” from his speech, which would in itself certainly be a somewhat idiosyncratic feature.
Phoneticians describe the dialect spoken in Wapakoneta, Ohio, where Armstrong was born and grew up – and where he lives to this day – as bordering the Inland North Dialect and the North Midland dialect areas. We listened to recordings of several speakers from these areas, including Armstrong’s parents, as well as Hillary Clinton and Michael Moore, and we consulted phoneticians, dialect surveys and other data, finally concluding that there is no tendency in this dialect to drop or slur words. The bottom line is that ‘General’ American, which is the colloquial name for this dialect, is widely represented in the media, the movies and American politics. It is considered to be very clear, easily understood speech.
Phonetics of the sentence
When we speak it is not just the words we use which count, but the way we say them – including the rhythms, pitches and stresses that we use. In our analysis we created a spectrogram of Armstrong’s statement in order to observe variations in pitch (See ‘Spectrogram’ illustrations at end of this document). Like any other speaker Armstrong’s voice will show variation in pitch and emphasis in order to create meaning. In the first part of the phrase ‘one small step for man’ you will hear from the recording that the pitch on the word ‘man’ is a rising pitch – that is to say, the voice goes up on the word ‘man’. In the second part of the phrase ‘one giant leap for mankind’ the voice goes down on ‘mankind’. This is known as contrastive emphasis, and speakers use this in order to signal to their hearers that they are contrasting two different entities or two different types of entity. For example, in ‘I don’t want tea, I want coffee’ a speaker might typically have a rising pitch on tea and a falling pitch on coffee – or the other way around. Whichever way a speaker does this they would be unlikely to have the same pitch on ‘coffee’ as on ‘tea’. There would be no point since the fact that this is intended to be a contrast between the two propositions, i.e. ‘tea’ on the one hand and ‘coffee’ on the other, would be lost to the hearer. Hence, in using one type of stress on ‘man’ and another on ‘mankind’ Armstrong was effectively contrasting ‘man’ with ‘mankind’. He had a very important reason for doing this. The word ‘man’ can mean the same thing as ‘mankind’, in expressions like ‘the history of man…’ or ‘man and nature’, etc. So, in using both words, ‘man’ and ‘mankind’, Armstrong clearly meant ‘man’ not in the sense of ‘mankind’ but in the sense of ‘a single person’. He used these different pitches of his voice to emphasise the significant difference in meaning between the two words. This clearly demonstrates that the word “man” was meant to signify “a man”, “a single person”, as opposed to “humankind” in general.
Several people have noticed an apparent pause between the word ‘for’ and ‘man’ in the phrase ‘one small step for [ ] man’ and have taken this to be indicative of the possibility of the article ‘a’ having been ‘lost’ in transmission. The length of this section is about 0.05 seconds. The difficulty with the lost ‘a’ theory here is that there is a very similar pause between ‘for’ and ‘mankind’ – of almost exactly the same length as that between ‘for’ and ‘man’. Mark Liberman and Garth Wiebe came to the same conclusion. These observations suggest that there is no space here for the missing “a”. The spectral activity on the spectrogram between “for” and “man” is not indicative of a ‘space’ for the word “a”, it is simply the residue of the rhotic /r/ at the end of the word “for” (see below for further discussion of ‘rhotic’).
Speech rate and spontaneity
During his preamble to stepping off the pad Armstrong speaks at between 150-160 words per minute whilst addressing the science teams in the backrooms with his description about the properties of the lunar surface. He then consciously slows down to a rate of more like 130 words per minute, as he acknowledges the historic significance of the next moment while he steps off the pad, and switches to addressing the world with his ‘One small step’ phrase. This more deliberate speech rate is about the same as that used by a radio or television newsreader delivering the news and one could be tempted into thinking that this careful enunciation of the words signified that Armstrong was simply reciting a prepared speech. However, we thought about this point carefully and concluded that the speech was not likely to have originated in the form of written language. Firstly, Armstrong would have slowed down his speech rate because he now had a different audience, a new audience – one he had not broadcast to before – namely, ‘the world’.
Aware that his message was being broadcast worldwide, he would probably have been very conscious of the fact that many of the people who were listening were not first language speakers of English. That was why, in our view, he would have slowed down his rate of speech.
Secondly, when we look at the structure of the phrase it begins with the word ‘That’s’. By ‘that’s’ here, Armstrong meant ‘the step that I’ve just taken’. Linguists refer to the use of ‘that’ in this context as deixis, a process whereby we use a word like ‘this’, ‘here’ or ‘now’ to give spatial or temporal context to our words. This suggests that although Armstrong may have thought about the phrasing beforehand, it only really came together at the last moment as he was stepping off the lunar module onto the moon. Notice also, that there is no clausal conjunction between the two parts of the sentence “That’s one small step for man”, and “one giant leap for mankind”. We felt that if the speech had been written for him in advance, the writer would perhaps have felt an inclination to use a more ‘elegant’ way of joining the two halves of the sentence, for example ‘but one giant leap for mankind’ or even ‘and one giant leap for mankind’. You can also see, from the film, that Armstrong is moving as he says this. When he says ‘that’s one small step’ he means ‘that step I’ve just taken’. He then moves to the left, and adds ‘one giant leap for mankind’. Both ‘small’ and ‘giant’ are emphasised exactly as we would expect them to be under the circumstances – again, we have contrastive emphasis between ‘small’ and ‘giant’. This strongly suggests that the speech was spontaneous rather than prepared or written out by someone else for Armstrong to say.
‘Small step’ and ‘giant leap’ in their historical context
In this context, it is also worth noting that expressions using “small step” and “giant leap” were being used by American writers and broadcasters throughout the 1960s to help capture the rapid technological advances of the period, and Armstrong seems to have borrowed and adapted them for his own broadcast. Using insights from a discipline called corpus linguistics we conducted searches of the language of the press and media of the day and came across countless references to elements of the famous phrase having been in use for at least 10 years prior to the first moon landing.
Thus, the Chicago Tribune of Christmas Day, 1960, said: “What the world needs today even more than a giant leap into space is a giant step toward peace”, while the New York Times of May 29, 1964 declaimed that the Saturn rocket test of the previous day “…was a welcome but small step on the long road to put a man on the moon and return him safely to earth”. As a last example, a headline in the Hartford Courant of 16 November, 1966, describing the touchdown of Gemini, read: “Pilots safe, Gemini Hailed for Giant Leap Toward Moon”. Hence, it is perhaps no surprise that Armstrong uttered the words he did as he stepped out of the lunar module. There was no great difficulty in making the connection between the two fairly common contextually relevant phrases of the period, ‘small step’ and ‘giant leap’. Armstrong would not have needed a speech writer, or any great thought, to have come up with this phrasing.
Stressed and unstressed words
Stress is used by speakers to highlight or emphasise meaning. Only some of the words in an utterance can be stressed. Some words are hardly ever stressed: ‘a’ is one such word. There are good reasons for this. ‘A’ is referred to in the grammar books as an ‘indeterminate article’ or – as linguists sometimes joke – an ‘indeterminate determiner’. Because it refers to something indeterminate, or unspecified, it is hardly ever stressed. In contrast, even though the word ‘I’ is also a very short word, it can nevertheless be stressed. Because ‘a’ is hardly ever stressed it is easy to miss – either when we say it ourselves or when other people say it. Although we have not carried out actual surveys of this point, we feel that it may not be that unusual for speakers to miss out the ‘a’ in normal running speech. Phonetically and phonologically the ‘a’ is also not that significant. Moreover, speech – like many other human systems – has inbuilt redundancy. In other words, if one element is lost another element will serve to keep the structure intact: thus, although we lose the ‘a’ we have the contrastive emphasis in the different pitches of ‘man’ and ‘mankind’: the information is there in Armstrong’s speech – all we have to do is listen for it.
In Armstrong’s dialect most speakers will pronounce the ‘r’ at the end of a word, such as ‘for’. Those dialects where final /r/ is pronounced are known as ‘rhotic’ dialects. There are a number of British, Irish and American rhotic dialects: for example, the New Jersey dialect is a rhotic dialect, but New York is not. Northern Ireland is, but the south east of England is not. The word ‘a’ in ‘one small step for a man’ would come after the rhotic ‘r’ and before the nasal ‘m’ in ‘man’. /r/ is a very special speech sound. Unlike many consonants the tongue doesn’t actually touch any of the other articulators when it makes the /r/ sound. Phoneticians refer to the tongue approximating the passive articulator. At this point the tip of the tongue is sitting vibrating behind the teeth in an area known as the alveolar region. It then moves down and forwards in order to facilitate the nasal /m/ in ‘man’, which is in any case a stressed word. It would be very easy to lose the ‘a’ under these circumstances.
Human performance factors
Another reason why it is possible to miss something as insignificant as the ‘a’ is to do with human performance factors. Speech is a system within a system (the human being) and that human being is a system within another system (society). We should bear in mind that Armstrong had endured a difficult landing on the moon, when there was a very real danger that the lunar module would run out of fuel. The astronauts then carried out various functions for about six hours before they opened the module and Armstrong stepped out. It is likely that the difficulties of landing, the anxiety about lift off, and the enormity of what he was having to do at that moment were a little more important than the niceties of speech elocution.
As speakers we all make small errors – of pronunciation, grammar or meaning, throughout every day of our lives. Yet as speakers we are often surprisingly naïve about our performance, failing to notice the many false starts, grammatical labyrinths, hesitations or mispronunciations. However, the main reason we are unaware of performance errors is that for the most part they don’t matter. Our hearers and listeners either don’t hear them or they adjust to them unconsciously: there will be other clues within the context which would tend to clarify or subliminally correct minor errors. As speakers and listeners we are programmed to ignore minor errors, to adjust for them, and to infer meaning from other sources, such as the context in which the words are uttered.
Clarity of Radio Transmission
In addition to examining the nature of Armstrong’s speech patterns we felt it was worthwhile commenting on the possibility of the “a” having originally been spoken on the Moon and having disappeared in the transmission en route from the Moon, as NASA has suggested. The transmissions did have a tendency to drop out and clip syllables occasionally – as can be heard in “giant leap” where the initial /dz/ phoneme of ‘giant’ sounds more like /d/.
A few days later, on the way back from the Moon, Lunar Module Pilot, Buzz Aldrin, the man closest to Armstrong when he spoke the words – and having heard them before transmission to Earth, repeats the line in his own broadcast back to Earth summing up the meaning of their mission from his perspective. In this speech he quotes the sentence in full. He does say ‘a man’, in contrast to Armstrong’s ‘man’.
However, we need to recall that Aldrin was making this speech under very different circumstances from Armstrong. He is simply talking: Armstrong was talking and doing when he spoke. The fact that Aldrin can be heard quite clearly saying “a man” indicates the possibility that there was little difficulty in the transmission of short, unstressed words in the broadcast process. In addition, there are several examples of Armstrong saying the word ‘a’ while he was on the moon. For example, in describing the moon’s surface just before stepping out of the lunar module, he says that it is ‘almost like a powder’, where the word ‘a’ is clearly transmitted.The recordings of Apollo 11 are now over 40 years old, and the audio has been through significant processing since then. Many researchers before us – including Shann Ford in 2006 – used the more heavily compressed mpeg sound files.
As far as we know we are the first researchers to have used an uncompressed digital audio wav file from a first generation 24-bit-precision 48 kHz recording made from the original magnetic tape recordings captured at the time in Houston. Of course, this audio still suffers from the original downlink compression, but despite these limitations, the words around the important sentence are still perfectly audible and would qualify for a “5x5” pilot check – meaning they are perfectly audible and intelligible. As a result of our investigation, we do not believe that the transmission processes or the state of the recording robbed the world of ‘a’ between ‘for’ and ‘man’.
In agreement with other researchers we also conclude from our own spectrogram analysis that there is no “a” in the Earth recording of the sentence where it ‘should’ be, and we feel it is unlikely that this word was lost during transmission from the Moon.
Having listened carefully to examples of both Armstrong’s speech and the speech of those who may have influenced him, such as his parents or his co‑pilot, there are no apparent prior difficulties or problems of an hereditary or environmental nature which could have caused Armstrong any difficulties in his use of language.
Taking into account the circumstances and context of the speech we conclude that for an unstressed “a” sitting between “for” and “man”, spoken by someone who had a large workload at the time, it is not surprising that the “a” might not have been enunciated. However, redundancy is an inbuilt feature of language and the contrastive pitch between ‘man’ and ‘mankind’ is a clear indicator that the intention to say ‘a’ was there.
As ordinary speakers of the language most speakers are unaware of their own performance limitations in circumstances where cognitive load and environmental stress are high. How would most of us have performed under circumstances where we had just landed a spacecraft which was low on fuel and were about to take off using a new engine which had not even been tested prior to that moment? We suggest perhaps not very well. Indeed, it is our view that the ‘complaint’ about the missing “a” reflects much more on those who have complained than on the man who, in a moment of great concentration on the epic task in hand, apparently omitted an insignificant part of speech. If nothing else, this so‑called omission shows us that the tasks Armstrong were then engaged in were occupying his attention much more closely than his precise words: surely exactly what we would expect of an astronaut at a new frontier of human space exploration?
This insignificant ‘error’ also adds weight to the belief that the speech was not rehearsed or composed by some ‘wordsmith’ beforehand. Subjectively, it is an apt and poetical pronouncement at a significant historical moment, uttered not by an historian but by a history maker. No doubt more will be written about this subject in the decades to come, but the intent of Armstrong’s original statement is not in dispute.
John Olsson & Christopher Riley, June 2009
About the authors:
John Olsson is the Director of the Forensic Linguistics Institute (http://www.thetext.co.uk ) and analyses both written and spoken language as part of his work for law enforcement and attorneys as well as corporate and private clients. He is the author of ‘Wordcrime’ (Continuum, 2009).
Christopher Riley is a former planetary scientist and works today as a writer, broadcaster, film maker and archivist. He has contributed to over 30 documentaries and books on Apollo, and is the author of the new Haynes guide: Apollo 11 – an owner’s workshop manual.
The authors will be presenting their work at the Cheltenham Science Festival on the Saturday 6th June 2009.
The authors wish to thank the National Aeronautics and Space Administration for their assistance in supplying the uncompressed Apollo 11 mission audio tracks which we used in this study. We would also like to thank David Woods, Editor of NASA’s Apollo Flight Journal, for providing further high-quality mission audio. Other archive for this analysis was provided by the Footagevault collection and the BBC.
1. Peter Shann Ford’s original Paper - http://www.collectspace.com/news/news-100306a.html and as a pdf download here.
2. David Beaver, Armstrong’s abbreviated article: the smoking gun?
3. David Beaver One small step backwards. http://itre.cis.upenn.edu/~myl/languagelog/archives/003630.html#more
4. David Beaver & Lisa Davidson First Korean on the moon! http://itre.cis.upenn.edu/~myl/languagelog/archives/003639.html#more
5. Mark Liberman One 75-millisecond step before a “man”. http://18.104.22.168/%7Emyl/languagelog/archives/003632.html
6. Mark Liberman, What Armstrong said http://22.214.171.124/%7Emyl/languagelog/archives/003645.html
7. Garth Wiebe, One small step for man reaffirmed http://www.wiebefamily.org/One_small_step_for_man_reaffirmed.htm
8. NASA’s Apollo Flight Journal - http://history.nasa.gov/ap11fj
9. NASA’s recently re-digitised 48 kHz, 24-bit-precision uncompressed digital audio wav file of Armstrong’s words.
10. Clip of the CBS recording of “I’ve got a secret” featuring Stephen and Viola Armstrong – Neil Armstrong’s parents, on the day their son was made an astronaut in 1962.http://www.youtube.com/watch?v=zd7eWKCOk-A
© William Labov, Dept of Linguistics, University of Pennsylvania
‘(Step) for man’ spectrogram – showing the rise in pitch (in blue) during the word ‘man’ (underlined section shows continuity of phoneme underlined)
‘(Leap) for mankind’ spectrogram – showing the drop in pitch (in blue) during the word ‘mankind’ (underlined section shows continuity of phoneme underlined)